// changelog
Every entry is written when the change ships, by the people who shipped it, and it is the same list the product shows inside your dashboard.
// the record
We updated our FAQ, comparisons, press, about, trust and help pages so they accurately describe how Magistry works, including how disputes are tracked and how trust-ramp thresholds apply. This removes unsupported claims and brings the copy in line with what the product actually does.
Video ads can now include a captions layer with plain, karaoke, or word-pop styles that stay perfectly synced to the voiceover and fit within each placement's safe zone. Loudness is also normalized consistently across every exported video.
Magistry now runs a weekly sweep of your Google Ads accounts to propose fixes for mined search-term negatives, settings drift, and missing placement exclusions. These proposals go through the same review and approval tools as any other action, so nothing changes without your say.
Magistry now watches each product's ad performance and automatically excludes it from Performance Max listing groups once it shows clear signs of losing money with no conversions. Paused products are automatically reconsidered after a price, landing page, or creative change, or after a set period.
Magistry now blocks automated theme changes that would hide compare-at pricing through stylesheets or custom CSS, and only confirms your site verification once both Search Console and Google Merchant Center can read it back.
You can now request XAF 3.2, XAF 4.0, or CSV ZIP finance exports through the API, check their status, and download the finished file once it's ready. Trial balance exports also accept an as-of date within the fiscal year.
You can now design and approve a studio ad even if there isn't a campaign for it yet. Once approved, the ad is held ready and placed automatically when the matching campaign is created.
When a competitor's top seller is something you don't carry, you can now click straight through to the importer with the product already prefilled. You can also import all flagged opportunities from the page header at once.
Magistry now automatically suggests new creative when a promotion is about to end, a product is about to launch, or an active campaign shows signs of ad fatigue. Each suggestion comes with the reasoning behind it, and you can review and approve it before it goes live.
If switching to a cheaper plan would push your projected overage past what that plan can bill, the downgrade is now refused up front and you're shown when your current plan renews. Overage above a plan's billing ceiling is also no longer run unbilled, so stores are suspended with a clear reason instead.
When you launch, Magistry now plans a first campaign for each advertising channel at your profile's daily budget cap, and creates it paused in Google for your review. A second confirmation after go-live switches it on, only once Google confirms the paused campaign matches your cap exactly.
The copilot can review your catalogue import and surface an audit verdict, with any override requiring your explicit approval before anything publishes. This adds a safety check to catch catalogue issues before they go live.
Before launch, Magistry now runs a set of checks across your policies, pages, delivery rates, theme, and shop details, and shows which ones pass or need a fix. A separate checklist highlights the handful of things only you can fix, like store address, cookie banner, checkout branding, and payment setup, with direct links to go fix them.
The launch page now shows Merchant Center item issues for each product, compared against sibling products and the rest of your catalogue. New stores also get sensible feed defaults automatically, so sale pricing and language settings are ready from day one.
A new conversational copilot walks you through launching your store one step at a time, proposing each action for your review. Sensitive steps like spend, account, publishing, and go-live always require your explicit approval, and you can set how much autonomy the copilot has.
The OpenAI product feed now checks each item against the latest spec, including ads eligibility, supported target countries, and pricing rules like valid GTIN check digits and sale price comparisons. Validation messages now list every missing field for an item instead of stopping at the first one.
You can now connect your product feed to Snapchat, Reddit, TikTok and OpenAI directly from the destination card, with no URL to copy or paste. Hit Send to register the feed and Check status to see the platform's processing result, including a clear reason and a Reconnect option if something is refused.
Merchants can now turn on and register their Pinterest product feed directly from the dashboard with a single button, instead of pasting a feed URL. The card shows registered, refused, or processing status per market, and automatically rolls back if registration fails.
You can now turn on features like adding new products directly from the agent screen in the mobile app, protected by Face ID, instead of hitting a dead end. Two new screens under More let you check every store you can access and see ad spend and top campaigns, while budget or bid changes still require approval in the Inbox.
When a Shopify payout is held, the health card now advises adding a second payment provider in Shopify as a fallback, helping you keep accepting payments while the hold is resolved.
Candidates without a price or image, and any candidate scored on fewer than two data sources, no longer show an inflated score — their score is now blank until real evidence exists. The candidate list also separates confirmed findings from live ones, ranks them accordingly, and shows how much evidence backs each score. Research lanes that have repeatedly found nothing useful for your store are automatically retired so effort focuses where it counts.
The customer-service inbox now shows one card per action and order, reflecting its live status (awaiting, running, blocked, or done) so you always see an accurate picture. Expired proposals no longer show an Approve button, and approving an action that is already running or completed is blocked to prevent duplicate work. Refunds that match the full amount of the original transaction are now issued in the correct currency.
When a broken product or collection page has no close live match, Magistry can now fall back to a relevant live collection — such as the product's category or its nearest parent collection — rather than leaving the 404 unrepaired. This fallback is opt-in per store and never points shoppers to low-traffic collections or generic pages like the home page or your full catalogue. All existing dead pages are re-evaluated immediately when you enable the setting, not just newly discovered ones.
A new audit findings card on the Ads dashboard surfaces open issues with your ad account so you can act on them without waiting for a support request. Each finding explains what was detected and what to do next. The panel is read-only and updates automatically as issues are resolved.
Your Brain now surfaces performance data for each studio ad asset — including spend, click-through rate, ROAS, hook rate, and hold rate pulled daily from Meta and Google. When an asset's engagement is declining, the Brain flags it as fatigued so you know which designs to refresh before they drag down campaign results.
You can now record a voice-over directly in the Music panel of the creative editor, using your brand's voice and language settings. The voice-over plays as a second audio track alongside your background music, which automatically ducks under the narration. Exported videos include both tracks mixed together.
You can now upload a separate logo for dark backgrounds in your Brand Kit, and Magistry will automatically use it whenever an ad creative has a near-black layer. You can also set a preferred voice and language for voice-over, so generated ads sound consistent with your brand from the start.
A new 'What Magistry repaired' card on the Visibility tab shows you the rewritten targets still live in your store, a live comparison of your newest product rewrites, and which store intent fixes have been applied and confirmed. Pages that haven't been checked yet are shown as 'not checked' rather than mismatched.
When a compliance hold is blocking products from your feed, the readiness panel now names the hold and shows how many products it is affecting. After you press Push now, the markets page also shows what the run actually did — including how many products were held, skipped, or deleted — so you are never left wondering why your feed looks empty.
You can now create Pinterest ads in a paused state — using an existing Pin or a new one — and update the title and description of a promoted Pin directly from Magistry. All Pinterest write actions are protected by a per-store permission switch, so nothing sends until your account is enabled. Archived keywords are handled correctly and will no longer appear as active targeting options.
A new timeline band appears on your marketing dashboard showing each planned action as a bar across the dates it covers — decide-by, measurement window, and verdict. You can click any bar to open the action card directly, and a dedicated Plan page lets you view the timeline at 2, 8, or 13 weeks and see what has changed over time.
When exporting a creative, you can now see all your connected Google and Meta campaigns listed alongside whether each one can accept the file — and if not, why. Add a headline and description for Meta, then send the asset as a draft ad creative using the same review-and-confirm flow you already know from social posts.
You can now propose a finished studio design as a draft creative on an existing Google or Meta campaign, complete with headline and description for Meta ads. The draft waits for your approval on the Ads page before anything is sent to the ad platform, so you stay in control of what goes live. Unsupported formats — such as Google video, incompatible sizes, or Meta ads missing required fields — are rejected immediately so nothing gets stuck mid-push.
Stores with rotating deals enabled now automatically cycle a set of products through real Shopify automatic discounts, each capped by order count and time window. A 21-day cooldown prevents the same product from appearing too frequently, and a live countdown is written to your storefront so shoppers always see an accurate deadline. EU stores are checked against the Omnibus price-history ceiling before any deal goes live.
When a product has no recorded purchase cost, Q4 readers now estimate its cost at one third of its current selling price so it can participate in assortment, feed label, and unit economics analysis. Your real cost always takes priority the moment one is on file, and nothing estimated is saved to your catalogue. Store margin and go-live checks continue to use only measured costs.
When a video creative goes through review, the system now examines frames at roughly one-third and two-thirds of the way through, not just the opening poster. This means pacing, on-screen text timing, and mid-clip moments are all considered before a creative is approved.
When a major new feature launches, you'll now see a compact card in the corner of your dashboard rather than a modal interrupting your work. The card never steals focus, and you can open the release, dismiss it for now, or turn off product update notices entirely. If you've already visited the new feature in the last 30 days, the card won't appear at all.
When you ask the AI to make, update, score, or export a video, a design card now appears directly in the conversation — no more hunting for the result in the studio. The card plays your video, lets you switch sizes, add or change library music, and download the version on screen, and it tells you when the studio has finished a newer version.
You can now set a daily cap and a lifetime allowance for ChatGPT Ads directly from Ads → Settings, giving you control over how much Magistry is authorised to spend on your behalf. Ask Magistry can also answer questions about your current ChatGPT Ads authority. When you reconnect an ads key, your conversions key is carried over automatically if the account hasn't changed.
When a design has performance data, you can now click "Make the next design" in the Results panel to start a follow-up. The composer shows you what the original said and how it performed, automatically groups the new design into the same test, and blocks any wording that overlaps too closely with the original so each variant stays distinct.
The Creative library and editor now show reach and engagement rate for every published post made from a design, broken down by network. When you run an A/B or multi-arm test, Magistry automatically calls a winner once one design has a 95% probability of being best and every design has reached enough accounts. Results are also available via the API and AI assistant.
Text layers in the design editor can now pull a quote, reviewer name, or star rating directly from your customers' reviews. The composer suggests your best eligible reviews automatically, and the editor's link drawer lets you search and pick any review by what customers wrote. If a review is later edited or removed, the design flags it so you can choose fresh words rather than silently showing outdated copy.
Holding Alt while a layer is selected now shows the exact distance to nearby layers or canvas edges, with a dashed guide line when they don't overlap. Dragging a layer between others snaps to equal spacing and displays the balanced gaps visually. Template and design thumbnails also animate on hover or focus so you can preview motion before committing.
You can now add a feed rule that withholds named secondary or lifestyle images from your product feed while leaving them untouched in your store gallery. This is useful when certain images carry claims that should not appear in ads but must stay on your storefront. The main product image is never affected.
Text layers in the creative studio can now be bound to live product or promotion fields — like a title, price, or discount — so your designs always reflect what's in your store. When something changes (a price update, a sale ending, a product selling out), you'll see a clear notice and can refresh the design or export as-is before anything goes live. The composer also detects matching text automatically and links it for you.
The History tab in the creative editor now includes an Activity view alongside Versions, showing every status change, AI request, render, comment, and social post derived from a design's renders — all in one timeline, newest first. You can filter, search, and export the full history as a CSV. Any team member or viewer with access to the design can see it.
You can now switch to comment mode on the design canvas to pin feedback to any spot, layer, or video moment. Comments are organised into threads that can be resolved, and you can ask the AI to apply a fix straight from the thread — the AI's edit appears on the canvas while the thread stays open for further discussion.
You can now ask Magistry to turn a single brief into a campaign set of 3–5 distinct concepts, each with its own angle, hook, and visual direction. A live progress bar in the creative library tracks each concept as it is composed, and the finished set is grouped together so you can compare and choose. The same workflow is available via chat and the API.
A new Brand Kit tab in the creative studio lets you review and set your brand colours, fonts, and other identity values field by field, alongside the evidence gathered from your website. A kit you set by hand is never silently overwritten by an automatic website reading, and the assistant can now read or update your brand kit directly from chat.
A full design studio inside Magistry. Start from your brand kit, let the AI propose directions, animate layers on a timeline, add music, and post it to your socials once you approve.
Magistry can now visit your store and extract your brand colours, fonts, logo, and tone of voice directly from your live pages — no manual setup required. Each value shows where it was found so you can review and adjust. Any brand details you've already entered by hand are kept as-is and the new reading is offered as a proposal instead.
When a post's image or video was made with an AI generator, Magistry now sends each platform's required AI content label on your behalf — TikTok, Instagram, and YouTube each receive their own native disclosure flag. Posts created without a generator are not affected. In the photo editor, any image produced by a photo tool now shows a before-and-after comparison with a wipe slider or side-by-side view, and you can restore the original as a normal undoable step.
Before a creative post goes live, Magistry now automatically checks prices, discounts, promotions, and any attested claims against your own store records. Issues that could block a post are flagged clearly and require your acknowledgement before publishing can continue. A check that could not fully run will never be shown as passing.
You can now animate text layers with a Typewriter effect that reveals letters one at a time, and photos with a Ken Burns effect that slowly pans and zooms the image inside its frame. Both presets are available in the animation picker when editing a creative, with sensible default durations set automatically.
You can now keep a generated take as a standalone design instead of applying it to your current one. Saved takes are grouped with their original in a new "In tests" view in the library, with each card labelled by what axis it differs on. This makes it easier to set up A/B tests directly from the creative editor.
You can now ask Magistry's chat, Brain automations, or the API to generate design takes and swap one in — the same way the design editor already does. Variations queue safely against the latest version, and a take that no longer fits a design that has moved on is refused rather than forced. Winning takes can also be placed into a test group so they run against each other automatically.
You can now ask for 2–4 takes on any saved creative, typing a direction or picking from a set of starter prompts. Each take previews live on the canvas in read-only mode, and a side-by-side overlay lets you compare every take from one request at once. Nothing changes in your design until you choose a take — and choosing counts as a single undoable edit.
When you ask for variations on a saved design, the AI now generates two to four distinct creative directions in a single request instead of one at a time. Each direction is validated before it reaches you, so only options that actually work and look different from each other are shown. You can browse the takes and adopt whichever one you like as a normal edit.
Common actions — text size, brand colour, bold, fill, entrance animation, alignment, AI edits, duplicate, and delete — now appear in a toolbar right beside whatever you have selected, so you spend less time hunting through the side panel. The toolbar automatically repositions above or below your selection to stay on screen, and works correctly even for rotated layers.
When you select three or more layers, you can now distribute them so every gap between neighbours is identical — horizontally or vertically. The outermost layers stay put and positions are rounded so gaps never drift as you keep editing. Works with undo and saves just like any other move.
A History tab in the creative editor now lists every saved version, grouped by day, showing who made each save (you, a teammate, the AI, or an automation) along with a live preview. You can filter by people or AI, page through older versions, and restore any version as a new editable draft without overwriting your current work.
Text, picture, and shape layers each appear in a distinct colour on the timeline, making it easier to read your composition at a glance. The tint carries through the bar, its entrance and exit handles, and the icon beside the layer name in both light and dark themes. Arrow-key nudges that were already at a boundary no longer leave unwanted undo steps.
The creative studio now has a full timeline showing every layer as a draggable bar you can slide, trim, and snap into place. Drag the bar to shift a layer's timing, drag an edge to trim its entrance or exit, and trimmed static layers gain a short fade automatically. Changes snap to a grid, the playhead, and neighbouring layers, and each drag lands as a single undo step.
The Brain can now queue a creative design request on your behalf, generating an on-brand draft project in your studio library in square, portrait, or story format. Nothing is published or sent — you review and publish the draft yourself. Up to 3 creative designs can be commissioned per day.
The compose drawer now opens with a 'What is it for?' step, where you pick from fourteen creative jobs such as spotlight, offer, how it works, event, or hiring. The choices are ordered for your kind of business, so a service business sees its most relevant jobs first. Picking a job alone is enough to get a focused result without writing a brief from scratch.
The AI compose panel now includes a catalogue picker so you can choose exactly which products or plans to include in your creative. Best sellers appear first, or you can search by name — each option shows today's live price and the compare-at price when available. The AI starter prompts also work for studios, software plans, and practices, not just shops with collections.
When you create or edit an ad creative, the designer now pulls in your store profile, your products with their live prices and compare-at prices, and any promotions running or starting in the next 45 days. Every price, discount and claim on the design is grounded in your actual catalogue — no more placeholder numbers or generic copy. You can also pin up to eight specific products to feature, and the designer works across any type of business without needing industry-specific templates.
You can now start designs, save versions, queue AI edits, export assets, and propose social posts through the Magistry API and via the AI chat assistant. A new `creative` scope covers all nine actions, so external integrations and MCP-connected tools have the same capabilities as the studio itself. Write operations verify store ownership before proceeding, so access is always enforced by your own permissions.
The Moments admin page now includes a Session time section that shows whether moments are reducing how long people spend per session, compared to their own activity over the prior four weeks. If median active time per session drops significantly for people shown a moment, the system automatically reduces how often moments can interrupt — and the page shows exactly what triggered that decision, or 'no verdict yet' while data is still being collected.
You can now choose from eight additional animation presets in a picker that plays each one on hover exactly as it will export, so there are no surprises. Press Cmd-K or / inside the editor to search every available action in one list, and type anything that doesn't match to send it as an AI design edit. Press ? at any time to see a full shortcut reference.
The creative editor now fills the entire viewport, giving you more room to work on your designs. Use the top-bar button or ⇧⌘\ to enter full-screen mode — the app header slides away and your preference is remembered across sessions. Press ⌘\ to hide both side panels, and use Esc to step back through modes one at a time.
You can now connect a YouTube channel from your Social Accounts page and approved video posts will be uploaded there automatically alongside your other networks. If YouTube locks an upload private you will see a note on the account, and the note clears once a video goes public. Photo posts no longer cause a YouTube target to fail permanently — each network only receives the file type it supports.
The moments console now counts reflex dismissals (triggered within 4 seconds of delivery) separately from deliberate ones, so you can tell apart automatic noise from real merchant decisions. It also shows the median time from when a moment was offered to when the goal was achieved, giving you a clearer sense of how quickly moments drive action.
The Moments admin view now includes a Mailed column so you can see exactly when an activation nudge email was delivered to a merchant. This timestamp is preserved independently of other activity, so it stays visible even after the merchant visits their dashboard.
Money figures throughout the ledger and reports now show cents so totals always foot, and all dates use a single consistent format. The trial balance labels each column group with Dr/Cr, cash and bank figures come with an opening-balance caveat, and a business that still needs setup sees its outstanding to-dos rather than a misleading clear state. The books shell now includes a command palette (Cmd/Ctrl+K), a quick-search shortcut, and phone-friendly stacked layouts with sticky columns and a real pager.
From the export drawer, you can now send a completed render straight to any social network your store has connected. Images and videos are matched to the networks that support them, and for video posts a cover image is included automatically. The post goes into your approval queue first, so you stay in control of what goes live and when.
You can now fine-tune any image layer in the creative editor with sliders for brightness, contrast, saturation, and blur, plus horizontal and vertical mirror toggles. A reset button brings the image back to its original appearance. These adjustments are locked alongside other style settings when a style lock is active.
You can now type instructions like "remove the background" in the creative editor and it will automatically queue the right photo tool on the image layer you named. The editor pauses while the tool runs and will not queue the same request twice, so you stay in control of your credits.
Four AI-powered tools are now available in the photo editor: remove a background, sharpen and upscale an image, erase objects with a brush, and expand a photo to fill its frame. Each edit costs credits, is charged only once even if you retry, and lands in your media library as a new undo step.
You can now give any scene its own soundtrack — generate a 3–60 second instrumental track using AI (billed in credits per started minute), upload your own MP3, M4A, or WAV file, or pick from your store's library. Gain, fade-in, fade-out, and start offset are adjustable in the Music tab or via AI chat commands, and the track is mixed into the final export at the same levels you hear in the preview.
You can now prevent Magistry from sharing your billing phone number with Google. When opted out, the number is excluded from both your published business contact and the verification text flow, so a personal mobile stays private.
When Magistry has a suggestion tied to a specific button or link on the page, the tip now appears as a small callout beside that control rather than as a generic corner card. If you navigate away before dismissing it, the tip moves to the corner so you never lose the message.
A small accent dot now appears on the navigation item linked to your store's top recommended moment, so you can spot where to go next without any interruption. The dot also appears on the category icon in the side dock before you open the menu. It disappears automatically once the goal is reached, you dismiss it, or the recommendation changes.
You can now open any creative project in a full editor with a live canvas, a layer stack you can rename, hide, lock, and reorder, and an inspector for layout, text, images, shapes, and motion. An AI chat lets you describe changes in plain language, your brand panel keeps colours and fonts on-hand, and an export drawer lets you download finished creatives. Every change is a single undo step, and if someone else saves while you are editing you will be asked how to handle it — your work is never overwritten.
When your profit report shows a cost that can't be calculated, Magistry can now use that as a signal to suggest the right next step — not just the page you're on. The suggestion still respects your usual limits so you won't see it more often than before.
When a goal you were shown in Moments is completed, you will now see a one-time toast notification confirming it closed and showing the evidence that got you there. Merchants who never saw the goal card are not notified. Undelivered milestone toasts expire after 7 days.
When you reach a meaningful milestone in setting up your store, a brief toast notification now appears to mark the moment. You can turn these off at any time from your guidance settings.
Approved creative scenes can now be exported as PNG, JPG, WebP, or MP4 files rendered by the same engine that powers the live preview — so what you see is exactly what you get. Still exports capture the scene at the poster frame; video exports render every frame at 30 fps. Scenes that fail validation, have missing assets, or contain text that overflows are rejected with per-layer feedback before a file is ever produced.
Costs has returned to Catalog and Expenses has returned to Insights, exactly where they were before the Finance section was introduced. The Finance group no longer appears in your store or workspace menu. All your existing navigation shortcuts and saved links continue to work.
Magistry now detects when your store is paused or billing-suspended and stops showing guided action cards that would fail to complete anyway. Four additional automation moments — store backup, refund adjustments, copy rewrites, and feed attributes — are now available to prompt you at the right time once your store is active.
Briefs you approve now produce real images and videos that are pushed directly to your Meta and Google campaigns — no manual export step needed. You can see each asset's rendering, held, or failed state right on the Briefs page, and approve or reject only affects assets that are still in the expected state. Meta video ads are created paused so you stay in control before anything goes live.
If you haven't submitted feedback or seen the rewards intro in 30 days, the intro card and button highlight will reappear as a gentle reminder. Each time it shows, the 30-day window resets, so it will resurface at most once per month.
The rewards intro card has been toned down to a cleaner, more neutral style. Gold highlights are kept where they matter — the coin, earned-credit notices, and the feedback pill — while the card itself no longer uses a gradient or animated sheen.
Costs is back under Catalog and Expenses under Insights, at their original addresses, and both read in English like the rest of the dashboard. Links saved while they sat under Finance still open the right page.
The feedback button now glows gold until you open it for the first time, when a card explains how many credits you can earn for leaving feedback. Any credits granted to you as a reward or goodwill gesture appear once as a '+N credits' notice inside the widget, so you always know what you've received.
You will now receive an email each time a feedback reward is granted — report rewards include the report name, what was actioned, and the credits awarded. Compensation emails explain what went wrong and include an apology, and are always delivered regardless of your notification preferences.
You can now earn free Magistry credits by leaving an app review, participating in giveaways, or signing up through an affiliate link. A card in your dashboard shows any active offer, and credits are applied automatically once your action is verified. Each reward can only be paid once per promotion, and unused credits carry forward to future billing.
Magistry can now send your purchase events directly to LinkedIn through their Conversions API, just like it does for other ad platforms. Once you connect your LinkedIn Ads account and Magistry creates a conversion rule for you, every purchase is forwarded automatically with deduplication against the LinkedIn Insight Tag. Stores connected before this feature was enabled will need to reconnect to grant the required permissions.
Product match proposals waiting for your review now appear as a dedicated lane in Pulse's "Today's triage" section, so you no longer need to visit a separate page to notice them. Each connected store shows how many proposals are waiting and how old the oldest one is, turning to a warning style when proposals have been waiting too long. You can approve all exact matches in one action straight from the lane.
Your product catalog can now be pushed directly to Meta, with delivery status visible on the Feed Health page. If Magistry is missing the required Meta permissions, the run is paused and flagged so you can grant access before it retries. Products pending Meta review are tracked across runs and updated automatically once Meta settles them.
Your product catalog can now be delivered directly to Meta. The feed health page shows each item's Meta review status, and if Magistry is missing the necessary permissions your catalog run is paused and labelled accordingly so you know exactly what to fix. Product availability is mapped correctly, and descriptions fall back to the product title when none is set.
Your public feed dashboard now shows tailored feed URLs for each ad platform — Pinterest, Microsoft, Snapchat, TikTok, and Reddit — so you can submit the right format directly without manual tweaks. Each platform fetch is tracked hourly, and empty feeds are flagged as failed so you always know your feed is healthy.
You can now connect Magistry to OpenAI Ads Manager by providing your SFTP credentials, which are stored securely per store. Once configured, your catalogue is packaged and delivered directly to OpenAI's server on each feed run. The channel is off by default — turn it on from the Feed Channels page when you're ready.
You can now push your product feed to Microsoft Merchant Center via the Content API, with per-product status reported back after each delivery run. Availability, pricing, condition, and product identifiers are mapped to Microsoft's format automatically. Choose between direct API delivery or import-from-file mode per store.
A new diagnosis view at Ads › Diagnosis breaks down every keyword and search term that spent budget over the last 30 days without generating a conversion — but only flags a line as waste when the math shows a lucky zero is unlikely. Each line shows the exact spend in your account currency and the date range so you can verify it straight in Google Ads.
A new Channels hub on the Feed page shows every connected destination in one place, with status, recent run history, any active issues, and a preview of five products as they will be sent. Each destination also gets its own detail page where you can review what was delivered and trigger a send — which is validated before it runs so you are never left with a silent failure.
The What's new page now shows totals for all time, the last 7 days, the last 30 days, and today, alongside a 12-week bar chart you can click to zoom into a single week. You can filter entries by kind, area, or time period, and search across all of them — results are paged 40 at a time so the page stays fast.
When a bulk product deletion is held for review, the alert now clearly states the date when Google will automatically remove those listings due to inactivity — so you know exactly how long you have to act. A separate critical alert is raised if the deadline is approaching and the hold is still unresolved, giving you two clear options to avoid unintended listing loss. Clearing the hold resolves both alerts immediately.
When you ask Otto about tax on an order, it can now see the tax amount charged and include it in its answer. Previously Otto had no access to tax data, so questions about VAT or sales tax would go unanswered. You no longer need to leave the conversation to look up tax figures manually.
The What's new page now opens with how many updates shipped in total, in the last 7 and 30 days and on the latest day, plus a chart of updates per week that you can click to see one week. Updates are grouped by day, search is always available, and you can filter by kind, area and period. The list loads in pages instead of one endless scroll.
Your catalog can now be shared with Pinterest using a dedicated feed endpoint that automatically delivers product images at the 1000×1500 pixel size Pinterest requires, with white padding applied where needed. Images in your existing Google feed and additional image columns are unchanged. No Pinterest campaigns are activated by this change.
Every candidate you or Magistry has approved now appears on the candidates page, along with a clear reason explaining what's holding it back — such as missing cost information, awaiting supplier images, or a pending onboarding run. The approval refusal message also now points you directly to Settings → Automations and the Growth Scout agent so you know exactly what to do next.
The ads health check now inspects every active Performance Max asset group for missing required assets, weak ad strength, disapproved creatives, and whether your Q4 core products are covered. Any issue found is surfaced by name so you know exactly what to fix before peak season.
The Accessibility report is no longer in the sidebar menu — you can now reach it as a linked card from the Compliance controls page, with a back link to return easily. The report itself has been refreshed with a KPI strip, a resolved-share bar, organised tables, and a collapsed manual checks list. It also remains accessible via the command palette.
When your products have issues in Google Merchant Center, Magistry now surfaces Google's own explanation and help link directly on the errors page. This means you can read what Google says is wrong and jump straight to their documentation without leaving Magistry.
Social post proposals that have been left unreviewed will now show a clear expiry notice so you always know which ones need attention before they lapse. Stale proposals are automatically flagged, keeping your social content pipeline tidy and up to date.
Before you approve an older social post proposal, you'll now see a clear notice explaining why it's flagged for retirement. This helps you make an informed decision rather than approving content that may no longer be relevant.
When a crawl finds the same SEO problem on a page you already acted on, Magistry now flags it so you know the fix did not take effect. The finding is marked for review and, where the option is enabled, automatically re-opened so it stays on your radar.
The feed diagnostics page now shows a clear line for each landing page problem, naming the cause and telling you exactly what to do next. If a page was found but not yet live in Merchant Center, you get a direct link to request a re-crawl. If a product page could not be found at all, it is named so you know where to look.
Draft orders and manually recorded orders can now be excluded from your forecast series and order rollup via a per-store setting. A shadow breakdown shows exactly which orders would be removed before you turn the switch on. The setting is off by default, so nothing changes until you opt in.
The What the Brain expects panel on your marketing dashboard has been rebuilt around an interactive fan chart. Each metric gets its own tab showing the last 13 weeks of actuals flowing into the forecast, shaped by how confident the Brain is. Hover over any week or use arrow keys to read the exact figures, and a certainty meter and trust ladder on the right explain what is driving the outlook.
When you connect a Google Merchant Center account, Magistry now creates your feed channel straight away so your products are ready to sync. Until the channel is active you'll be guided to the feed setup page rather than landing on an empty state. No extra steps are needed from you to get started.
Each review source on your Reputation page now displays a clear status — such as connected, credentials missing, or paused — along with the right next step to take. Failed reads are shown as unread so nothing is silently lost, and spam reviews are excluded from your mention totals.
When a support ticket is closed, the notification message now includes a Rate this ticket link that takes customers straight to the rating panel without an extra click. Customers who do not return to the ticket page will no longer be prompted separately. No changes to the rating scale or survey wording.
The product page now loads revenue, net profit, and ROAS figures as 30-day totals rather than paging through individual variant rows, so numbers appear faster and more accurately. The product details drawer now displays these totals labelled with their time window and formatted in your store's own currency instead of a fixed euro sign.
You can now approve every exact-match link proposal across your workspace in a single action, instead of confirming them one by one. The links review page also shows how long the oldest open proposal in each tier has been waiting, so you can prioritise your review queue at a glance. If the age cannot be retrieved, the page displays 'age unknown' rather than leaving the field blank.
You can now approve every exact-match link proposal for your workspace at once instead of confirming them one by one. The linking queue also now shows how long the oldest proposal in each match tier has been waiting, so you can see at a glance where reviews are overdue. If the age cannot be loaded, the queue shows 'age unknown' rather than leaving the field blank.
You can now delete a block that was generated by your own feed settings, directly from the feed page. This gives you more control over what appears in your feed without needing to contact support.
You can now enable or disable the Omnibus Directive price history display directly from your compliance controls page, without contacting support. The switch takes effect immediately across your storefront.
When you reconnect your Pinterest account, your chosen ad account is carried over automatically as long as the new login can still see it. The catalog panel now checks your granted permissions before you try to use it, and shows a clear message with a reconnect link if catalog access is missing or was connected before permission tracking was added.
Your scheduled Black Friday promotion will now be announced to Google Merchant Center up to 7 days before it begins, with the sale price and exact sale window included on each product. Shoppers searching on Google will see your discounted price as soon as the promotion goes live, with no manual feed update needed.
The Settlements page now shows each open refund alongside its current status and a direct link into the relevant admin screen to resolve it. Refunds that have already been paid but were sitting in a parked state are now automatically closed out without moving any money. Stores on paused or suspended billing plans are included so no refund falls through the cracks.
The Fiscado API endpoints now expose orders, returns and invoice lines from bol.com and Allegro channels, giving your accounting integration the data it needs for VAT and OSS reporting. Each order includes the delivery country so cross-border obligations are correctly attributed, and Allegro invoice lines carry unit price, total and line identifier. Connect your Fiscado token to start pulling this data today.
Your marketing dashboard now displays forecast tiers — showing projected performance ranges and the reasons a forecast may have stopped — so you can plan spend with more confidence. You can also leave a thumbs up or down on any forecast to help improve future predictions. The feature is on by default for all stores.
The voice screen now shows the orb front and centre with three large round buttons — Type, Talk, and End — so it is clear what you can do at a glance. A single End button replaces the two separate close buttons that were there before. Text you speak appears centred beneath the orb, and status messages are written in plain sentence case.
The Data Processing Agreement Annex A now discloses how pooled forecast-calibration data is handled, giving you a clearer picture of your data rights and obligations. The trust band on the site has also been updated to reflect this. Review the updated DPA to stay informed about how your data is processed.
If your OpenAI API key is rejected by OpenAI, Magistry now records it and surfaces it on your workspace health card so you can see exactly which key needs replacing. Previously, a refused key could silently fall back to the platform key without any visible warning to you. Swapping in a new key clears the alert immediately.
The assistant conversation now automatically scrolls to keep the newest message in view above the keyboard, and snaps back to the bottom each time you send a new message. Approval requests now lead with a clear badge so you can spot them at a glance, alongside the Approve and Reject actions.
When an AI agent connected via MCP proposes automation changes or declares an intent, Magistry now presents a plan card that requires your explicit approval before anything is applied. Auto-apply cannot bypass this step, so no automation setting changes without a human in the loop. Disabling automation remains immediate as before.
When connecting Meta Ads, you now pick the Facebook Page that will be shown on your ads — complete with follower count and a live ad preview. The chosen Page is also visible in your Ads settings so you always know which Page is active.
The assistant on mobile now looks and works like the chat apps you already know: your messages appear in bubbles, answers span the full width with their sources tucked above, proposals show as approval cards with Approve front and centre, and multiple-choice questions are tappable cards whose selection is sent automatically. A single floating composer with voice input and a send/stop button stays with you throughout.
The draft review screen now shows the order match, detected intent, key facts, tracking status, supplier verdict, ETA statement, and suggested action that the agent relied on when writing each draft. If a check was not run for a particular draft, that is stated clearly rather than hidden. This makes it easier to spot why a draft was written the way it was and catch anything the agent may have got wrong.
When you install Magistry from the Shopify App Store without an existing account, you now see a welcome experience directly inside Shopify admin: an explanation of what Magistry does, a step-by-step guide, and buttons to create a free account or link an existing one. Once your store is connected, a Getting started checklist helps you complete your setup. The sign-up page also shows which store you are installing.
AliExpress trend data was silently returning no results for an extended period due to a page layout change — this is now fixed and results are flowing again. You can now target up to two markets when running a research brief, and the brief form shows you the estimated credit cost before you commit to a run.
When reviewing a customer service draft, you can now see the relevant order — including tracking and per-item stock and cost — directly beside the reply panel, so you have the context you need without switching screens. The matched order loads automatically even when it predates the customer's most recent orders.
The workspace overview now includes a period picker so you can view totals, per-store revenue, and top-performer insights for Today, Yesterday, the last 7 days, or the last 30 days. Your selection is saved in the URL, making it easy to bookmark or share a specific view. All times use UTC dates, matching the way your KPIs are calculated.
A core product that has a disapproved offer can now pass the feed readiness check if it has an approved Shopping Ads offer in every market you are targeting. Stores with no targeted markets, or no core product, continue to follow the previous strict rule.
When an hourly spend alert fires, it now includes a breakdown of AI usage by purpose (such as CS agent, marketing brain, or Q4 flows) for the hours that triggered it. This lets you immediately see what drove the spike instead of having to investigate manually. If the spend came from outside known call sites, the alert says so explicitly.
Your strategy dashboard now surfaces plans that have expired and goals that cannot be measured, rather than silently omitting them. Expired strategies show their closed status, and unmeasurable goals display the reason they could not be tracked. This gives you a clearer picture of where your ad strategy stands.
A new panel on the Costs page lists your clone store pairs and any staged cost proposals, making it easy to spot where costs differ or are missing between linked stores. No costs are written automatically — you stay in control of any changes. Health checks now flag foreign numbers, different cost bases, and uncosted items beside their twin so nothing slips through unnoticed.
Magistry can now look up a product's supplier page, confirm it with a vision check, and pull in Tier-B cost estimates and size measurements automatically. You can review the supplier's confidence verdict and your store's measured cost accuracy band directly on the Costs page. Size data is only imported when a unit is clearly stated — ranges and ambiguous tables are never guessed.
Magistry now detects when your storefront pixel has gone quiet for more than 24 hours while orders are still coming in, and raises an alert so you can act before attribution gaps grow. You can also see a breakdown of funnel coverage by step — cart, checkout, and purchase — showing where measured data, modelled estimates, and missing signals each stand for any given day. This makes it easier to spot and fix tracking gaps before they affect your reporting.
Packshot creatives are now prioritised at the front of additional product images so the 10-image limit no longer drops a creative in favour of a Shopify image. Lifestyle creatives are mapped to the dedicated lifestyle image field, and hero creatives remain the primary image. Each image URL appears only once in the feed.
Once your last viable Christmas order date has passed, Magistry will no longer raise or resume ad budgets until 25 December — preventing spend from stepping up when orders can no longer arrive in time. The cut-off date is calculated per store and market, so you are protected automatically without any manual action needed.
You can now see a tailored Black Friday and Cyber Monday peak plan directly in your profit dashboard, giving you a clear view of what Magistry is preparing for your store during the Q4 rush. A pre-flight check runs automatically before any peak-period actions are taken, so you get early warnings if anything needs your attention. The plan runs in a safe dry-run mode first, so nothing changes until everything looks right.
You can now see at a glance whether your Meta, Pinterest, and Microsoft ad channels are ready to spend, along with the specific blocker and who owns it if something is holding a channel back. A spend-up gate enforces this verdict automatically so budget is never pushed to a channel that isn't ready. The full picture is available on your Performance & Profit dashboard.
Magistry now injects JSON-LD structured data into your storefront so that AI-powered search engines and crawlers can better understand your products. Breadcrumb markup is always included, and full product details are added when your store is set to complete mode. You can control the mode through your store settings.
A new review queue on the Creatives page lets you approve candidate images as your main or an extra product image, decline ones that don't fit, or approve everything in bulk with one click. A publish switch controls whether approved creatives go live, and a performance panel shows how your approved images are affecting click-through rates over time.
The onboarding checklist no longer requires Google Ads, ad spend, or a first sale before you can activate the Brain. The Brain can apply feed and SEO improvements straight away, and the wizard opens in approval mode so you stay in control of every action.
When a dispatch decision is held in a quarantine window, the approval card now clearly shows when it is scheduled to run and that dismissing it will abort the action before it executes. Previously, quarantined proposals looked identical to any other open proposal, with no indication of the pending countdown. This makes it easier to act decisively before an automated decision goes live.
The SEO findings page now groups issues by category (demand, on-page, technical, other engines) with tiles you can tap to filter the queue. Each finding opens a detail drawer showing the evidence behind it, with shortcuts to fix it using Magistry, open the affected page, or dismiss it permanently. Dismissed findings move to a separate view where you can restore them any time.
When you connect a Google Ads account, Magistry now automatically checks and enables auto-tagging on your behalf, and links your Ads account to your Google Merchant Center — so your campaigns and product feeds are properly connected from the start. You can control auto-tagging per store using the new switch on the Ads dashboard. The Ads-to-Merchant Center link is detected and accepted automatically, so you spend less time on manual setup.
The break-even ROAS figure shown on the Profit page now reflects your store's calibrated margin settings rather than a simple fee-only estimate, giving you a more accurate target for your ad spend. A fee-only figure is still shown alongside it for reference. If your margin hasn't been calibrated yet, the break-even ROAS will show as unknown.
Each open policy finding in your compliance view now displays a clear path to resolution — whether that's a suggested policy update, an action you need to take, or a platform-level change. You no longer have to guess what to do next when a finding appears.
Compliance-recommended theme edits now appear on the Theme Edits page grouped by rule, each showing a line-by-line diff of exactly what will change in your live theme. You must open the diff and tick your approval before any fix is deployed, and the system will refuse to apply a patch if your theme has changed in the meantime. Edits waiting on you are labelled "Waiting for your approval".
Connecting your Meta Ads account now automatically enables the Meta Pixel on your storefront, server-side Conversions API mirroring, and catalog syncing — no separate tokens needed. Events are consent-gated and deduplicated between browser and server. You can verify everything is working with the new 'Send test event' button, which fires a test purchase and shows you Meta's confirmation.
When a supplier stock-out is detected, affected products are now automatically paused in your shopping feed so you stop advertising items you cannot fulfil. Once the supplier confirms the product is back in stock, it is restored in the feed exactly as it was before. This means fewer cancelled orders and wasted ad spend during supply disruptions.
The Brain previously relied on market-wide transit estimates to decide when to stop scaling spend for Christmas demand. It now uses cut-off dates measured from your store's own fulfilment history where available, so the Last Order rule fires at the right time for your operation rather than a fleet average.
Magistry now automatically gathers competitor pricing from Google Shopping each morning so you can see how your prices compare to the market. Each product shows the median market price alongside the upper and lower range, with sources matched by image and link for accuracy. Where a reliable market price cannot be found, you will see a clear reason instead of a blank.
Magistry can now register and maintain a Pinterest catalog feed directly from your existing feed channel, using the channel's country, locale, and currency settings. The feed channel page shows the current processing status returned from Pinterest. Note: Pinterest catalog sales campaigns are not yet available to create while catalog processing is being established.
Magistry now reviews each active Google campaign once per day and increases its budget when ROAS is at or above break-even with enough measured purchases to be confident — up to +30% when performance is strong. Budgets step back down if a loss is proven at high confidence, and every decision is visible on the Profit page so you can see exactly what changed and why.
Shoppers now see a "Delivered within N days" message directly below the Add to Cart button on product pages, giving them a clear delivery expectation before they buy. The number of days is calculated from your store's measured delivery performance. This experiment runs automatically with no setup required on your end.
The changelog now opens showing new and improved entries first, so the latest features are front and centre. Fixed entries are still one click away under the Fixed and Everything tabs.
You can now see daily and weekly profit breakdowns in the Performance section of your dashboard. Tax is captured from your Shopify orders so your profit figures reflect real costs. Where profit or ROAS cannot be calculated — for example if spend data is missing — the value shows as blank rather than zero, so you always know when a number is genuinely unavailable.
You can now view a per-page, per-month breakdown of accessibility issues found on your storefront, including which WCAG 2.2 A/AA criteria require manual review. Issues are marked resolved only when a later full scan confirms they are gone, and automated fixes are counted separately. Download the report as CSV or PDF directly from the Accessibility page.
The Ads page now shows a card summarising the Google AI settings on each of your campaigns, including AI Max, generated text, final URL expansion, and brand exclusions. This is a read-only view so you can quickly check what Google has enabled without leaving Magistry. Note that AI Overviews and AI Mode placements are not reported separately by Google and will not appear here.
You can now see how ready your store is to be found and used by AI shopping agents, with a dedicated section on the AI Visibility page. The readiness score highlights what you have covered and what to improve so agents can act on your behalf or recommend your products accurately.
You can now jump straight to your optimization plan, support tickets, reviews, or campaigns in Magistry by clicking the relevant action link inside the Shopify admin. Magistry recognises your store automatically and takes you to the right page in a new tab. No extra sign-in or copy-pasting needed.
When your store has the AI disclosure setting turned on, the required disclosure line now also appears on AI-composed outbound emails and social media direct message auto-replies — not just on replies a person reviews before sending. Messages written by a person are never affected, and turning the setting off keeps messages byte-identical to before.
A new AI Output page is now accessible from the navigation, command palette, and help system. You can review and manage how AI-generated content is disclosed to your customers, and save your disclosure preferences directly from the page.
You can now tell Magistry to automatically approve a specific action on a store every time it comes up, without you having to review it each time. Set a daily limit of up to 50 approvals per day, and the rule runs silently in the background until you turn it off. An undo button and a 30-day inactivity flag in your inbox keep you in control.
A new AI Output page shows you how many outbound customer-service messages and social DMs were written by AI in the last 30 days, how many carried an AI-disclosure line, and the full history of any changes to that setting. You can turn the disclosure line on or off from this page, and switching it off requires a written reason that is saved for your records.
Magistry now checks your store's robots.txt file and raises an opportunity when it finds rules that block AI-powered answer engines from reading your pages. Blocking these agents means your store may not appear in AI-generated shopping recommendations or cited answers. The opportunity clears automatically once your robots.txt allows the relevant crawlers through.
The Magistry app is now offered for Android (13+) alongside iOS — find the Google Play badge in the site footer, on the mobile page, and in the dashboard. Android users will see a "Get the app" prompt in the dashboard linking to Google Play, and the smart redirect at magistry.io/get-app sends each phone to the right store automatically. Note: the Play Store listing is pending approval and may not be publicly available yet.
On Android, pressing back while signing in now preserves your email address so you don't have to retype it (your password is never kept). In Settings, the amount you entered in the Money line sheet is also held until you save or discard it.
On Android, pressing the back button now behaves predictably on every screen — modals close, forms stay put, and you return to the right place. If you start typing a reply or feedback and press back, your words are waiting for you when you return.
When something goes wrong in the mobile app, the error message now includes a clear action — like "Check your connection", "Sign in again", or "Wait a minute" — so you always know the next step without guessing. This is especially helpful when opening the app from a push notification.
When Magistry cannot load a page on mobile, you now see a clear message explaining the problem — whether it is a sign-in issue, a permissions issue, or something on our end — along with a prompt to pull down and refresh. Active automation runs also announce their status to screen readers on iOS.
If a screen runs into a problem, only that screen shows an error — the rest of the app stays usable. You can tap "Try again" to reload just the failed screen, or "Go to Inbox" to return to a working state straight away.
When biometric authentication isn't set up on your device, you'll see the reason and a direct link to open your device settings. If you're locked out, you'll see how to unlock without losing your saved enrollment. If you cancel the prompt, a clear message confirms nothing was changed and keeps the payment sheet open so you can try again.
On Android 13 and above, Magistry now asks for notification permission at the moment you enable push notifications in Settings, not on first launch. If you permanently deny the permission, you'll see a clear prompt with a link to your device settings. Permissions revoked in system settings are automatically synced so you won't receive alerts you didn't agree to.
On Android 13 and later, the app now asks for notification permission only when you turn on the push switch in Settings — not on first launch. If you later revoke permission in your phone's system settings, the app detects this when you return and updates your preferences automatically; re-granting restores them. Money approvals and critical alerts always appear as heads-up notifications.
The theme go-live page is now split into three focused sections — a library showing each theme's sync status, a preflight checklist with a before-and-after preview, and a run panel with step-by-step instructions and a live progress tracker. A capacity warning blocks publishing when you've hit Shopify's 20-theme limit, and you can deep-link to a specific theme with ?theme= in the URL.
If you duplicated your GitHub-connected theme and published the duplicate, Magistry was editing a theme your shoppers never saw. The new Go live page lets you pick any theme in your Shopify library and push it live through your repository — on a clean branch, so nothing changes for shoppers until you publish and the old theme stays in your library as a fallback.
A new Decisions inbox collects every open Brain proposal for your store so you can approve or dismiss them without hunting across different pages. Proposals that are ready to run are grouped under Needs you with bulk approve and dismiss controls per domain, while proposals that cannot proceed are listed separately. Other pending approvals link directly to the relevant page.
You can now see a breakdown of orders that arrived via AI referrers such as ChatGPT, with revenue grouped by engine, unknown AI source, or non-AI traffic. Attribution uses your journey data first, with pixel data as a fallback.
A new AI Visibility page shows how ready your store is to be discovered and cited by AI shopping engines, broken down into a weighted score across five areas. You can see AI-referred orders per engine alongside an untracked lower-bound count, and each part of the score links directly to a fix. Unmeasured areas are left out of the score rather than dragging it down unfairly.
Tools that create refunds, issue gift cards, or cancel orders are no longer accessible to third-party apps connecting via OAuth — they remain available only to your own API keys. This keeps your store's financial actions under your direct control. Tool descriptions visible in MCP-compatible clients have also been updated to be factual and neutral.
Every Magistry tool exposed via the MCP API now carries a display title and fully typed parameters, so connector directories like Claude and ChatGPT can list and describe them properly. Read-only and data-modifying tools are also correctly annotated, meaning AI assistants can apply the right safety checks before calling them. If you connect Magistry to an MCP-compatible AI client, you will see named, well-described tools instead of blank or flagged entries.
The store switcher on mobile now loads a capped first page of your stores and clearly tells you when there are more stores beyond what's shown. This keeps the switcher fast to open while making sure you always know if your full list isn't visible.
While a sign-in is in progress, the sign-in buttons now show a hint explaining why they are disabled. When submitting feedback, the Send and Reply buttons show a busy state with a hint, and any refusal messages are announced aloud by screen readers on both iOS and Android.
While your billing details are saving, the Save and Cancel buttons now show a hint so you know why they are temporarily unavailable. The delete account flow explains that your email address must be typed to confirm. If you try to save the same rejected amount more than once, the message updates to show how many times it has been refused and is announced to screen readers.
Screen reader users on mobile can now hear the full context when reviewing inbox outcomes: the snackbar announces the queued amount alongside 'Will send when online' and the option to undo. Each decision row speaks its amount and warns that Face ID is required when approving a money movement. Swiping or using the rotor to approve while offline now opens a clear explanation instead of silently doing nothing.
After a screen finishes loading, VoiceOver and TalkBack now announce what appeared — such as your inbox items or an empty state message — instead of staying silent. This applies across the main screens including Inbox, Home, Brain, Feedback, and the store switcher. Error and offline states already spoke and continue to do so.
When a session was revoked on another device, the sign-in screen now explains why you've been signed out so you're not left wondering. Malformed or unsafe links are rejected rather than silently failing, and unknown destinations now offer a clear route back to your Inbox.
When you revert an approval on mobile, a brief window appears letting you cancel that action before it takes effect. Tapping Undo during this window stops the revert immediately, leaving the original approval intact. No changes are saved until the window closes.
A 'Report this answer' option now appears beneath every AI-generated response in both chat and voice, letting you flag a problematic answer without leaving the app. Reports are sent automatically, and if you're offline they're saved on your device and submitted the next time the AI responds. Error messages and system turns do not carry the option.
When you remove a device from your account on the web, that device is now signed out at the same time — it can no longer receive push notifications or stay logged in. Previously, revoking a device only removed its push token, leaving the session active.
The push notifications switch in mobile settings now shows a clear reason when it cannot be changed — for example, because your preference is still loading, failed to save, or you are offline. Previously the switch was silently disabled with no explanation.
When your device is offline or data is still loading, the app now explains what you're seeing instead of showing blank values or misleading messages. The queue, feedback threads, voice assistant, and settings screens all keep their content visible under a status banner rather than replacing it with an error. Voice will not start listening when there is nowhere to send your input, and devices without a speech recogniser no longer show an unusable button.
You can now sign out of all active mobile sessions in one step — every device is logged out, push notifications are revoked, and biometric keys are cleared immediately. To protect against repeated attempts, each device and account now has its own request rate limit, so you'll see a clear error if a limit is hit rather than silent failures.
When you sign out of the Magistry mobile app, all other sessions tied to your account are also ended automatically. This means you no longer need to sign out from each device separately to fully secure your account.
When you remove a device from the web dashboard, it stays removed — the mobile app can no longer silently resurrect that session. A fresh sign-in on the device will still work as normal. Mobile API requests that exceed the per-account rate limit now receive a clear retry signal so the app can back off gracefully.
If Otto's reply stops arriving mid-stream, the partial answer is kept on screen with a clear message that it stopped rather than disappearing. A timeout no longer disables the Send button — it tells you that sending again starts a new request. If a reply failed before Otto could confirm it, the full answer is quietly fetched and shown, so nothing is lost.
When you turn on Reduce Motion in your phone's accessibility settings, the app now responds: navigation transitions, the biometric prompt, and the Money line sheet all cross-fade instead of sliding, and Otto's transcript jumps to new content without scrolling. This makes the app more comfortable for users sensitive to motion.
When you run or pause an agent from the mobile app, you will now be prompted for biometric authentication showing the agent and store name before the action takes effect. Cancelled checks are safely ignored and no action is taken. All agent actions are recorded in your audit log before they execute.
If Otto stops receiving a reply mid-stream, your partial answer is kept on screen with a note that it stopped arriving rather than disappearing. A full timeout now tells you clearly that asking again sends a second request, and a status check that takes too long explains itself instead of spinning. Tapping Send twice quickly now sends one message, not two.
You can now roll back an approved marketing action directly from the mobile app. The revert requires biometric confirmation, checks whether the action is eligible before attempting anything, and shows the updated status immediately after. If the rollback could not be submitted, the app tells you nothing changed rather than leaving you guessing.
If your Microsoft Ads, Reddit Ads, TikTok Ads, or OpenAI Ads connection lapses or expires, the dashboard's connection banner now shows a reconnect card so you can fix it straight away. Previously these platforms were not checked, so a lapsed token would silently stop your ad syncs with no visible alert. Microsoft Ads tokens are also kept up to date automatically after each refresh, reducing how often you need to reconnect.
The Inbox, decision detail, Otto, Home/Today, sign-in, and tab bar now all scale correctly when you use your device's larger text accessibility settings. Charts, badges, and stat rings resize without clipping or cutting off labels. All touch targets remain reachable at any font size.
When Magistry handles your ad platform tracking, the Pinterest Tag and OpenAI Ads pixel are now served automatically — no need to paste a second tag yourself. Purchase events are deduplicated so each platform sees only one signal, and nothing fires until a marketing-consent signal is received.
When you tap Approve or Reject on a push notification, the app now opens so you can confirm with biometrics or a second tap before the ruling is recorded. Only Snooze still works silently from the lock screen. This means accidental or unauthorised lock-screen rulings can no longer take effect.
When you choose Magistry-managed tracking for OpenAI or Pinterest, purchase events are now sent automatically as server-side conversions — no extra setup required on your end. For OpenAI, Magistry adopts or creates your pixel and wires up the order event; for Pinterest, your selected ad account becomes the conversions destination. If you've set a platform to self-managed or off, Magistry stays out of the way entirely.
Every ad platform you have connected now appears on the Tracking page with a live status showing whether your sales are reaching it — not tracked, setting up, live, quiet, failing, or off. You can also see and change how Magistry handles tracking for each platform, and new connections land here instead of Ads settings.
You can now generate a stable, shareable feed URL for each channel (Snapchat, Pinterest, TikTok, Meta, Microsoft, Reddit, and others) directly from Feed Channels settings. Platforms that pull your product feed on a schedule can use this URL to fetch your latest published feed. Toggle it on or off per channel at any time — disabled or unrecognised URLs return a 404.
When you connect an ads API key, your campaigns now sync straight away instead of waiting until the overnight refresh. You no longer need to sit until the next morning to see your first campaign data.
The public API reference has been updated to accurately describe risk tiers (read, auto_write, and propose_write), correct the status returned for proposed writes, fix the curl examples to use field names the API accepts, and replace the deprecated DCR flow with CIMD. Webhook support (CloudEvents and ES256 Standard Webhooks) and the 365-day versioning policy are now documented as well.
When your Merchant Center account is suspended for an account-level policy violation such as Misrepresentation, the specific policy name now appears on the suspended account banner, the recovery header, and above the review request panel. This makes it clear that the suspension was decided on that account-level finding — not on individual products — and points you directly to the right recovery step.
When pausing an agent from its detail screen, you are now asked to confirm before the change takes effect. Held states are described in plain language so it is clear why an agent is not active. If the pause or resume request is refused, the agent reverts to its previous state automatically.
When tracking is turned off for a store, return shipments are no longer polled or registered with the carrier aggregator. Stores with tracking disabled will have their returns escalated at the deadline and reported separately, so nothing falls through the cracks silently.
The status page now shows real-time platform health instead of a hardcoded 'operational' message, and any untracked component is shown as 'Not measured' rather than falsely green. Pricing overage rates and agent pause behaviour now match what you are actually billed. Two new pages — one for mobile commerce and one for the Standard plan — are available from the main navigation.
When the biometric authentication sheet is open on your mobile device, screenshots and screen recordings are automatically blocked to protect your credentials. This safeguard lifts as soon as the sheet is dismissed.
The mobile queue is now an Inbox split into three tabs: Needs you (items waiting for your decision, shown in your badge count), Ideas (research candidates ranked by value), and Done (last 7 days of sent, running, completed, failed, and reverted actions). Non-money approvals wait 5 seconds under an Undo banner before sending, and actions queued while offline are delivered automatically when you reconnect. Money decisions are sent immediately after biometric confirmation and cannot be undone.
The Magistry marketing site footer now shows Apple's official App Store badge so visitors can find the iPhone app easily. iPhone users also see a dismissible suggestion card after they engage with the page, which stays out of the way of the signup bar and respects any consent preferences.
When bol marks your offers as not for sale, the dark-listing card now lists the top three reasons — such as store closed or account blocked — along with how many of your offers are affected. If no fresh data is available the previous message is shown unchanged.
Tapping 'Run' on an agent in the mobile app now triggers a real run, just like starting one from the dashboard. You'll see the run status update to pending, started, or a refusal message from the server. The agent detail screen also properly shows loading and offline states instead of a blank page.
Your AI model spend on the Brain and System screens is now displayed in credits instead of dollars, matching how your billing ledger tracks it. If a section of the Brain screen fails to load, the rest still appears and the affected section is clearly marked rather than blocking the whole screen. Lists on the Brain screen now page through all results instead of cutting off.
When Magistry cannot fetch your mailbox, the alert card now names the exact setting and button to use — Update credentials, or reconnect via Microsoft or Google — and shows how long the issue has been happening. Alerts for removed mailboxes are automatically closed, and repeated failed login attempts are throttled so your inbox is not hammered while you make the fix.
When an item needs your approval, you'll receive a push notification at the 12-hour and 1-hour marks before it expires — complete with the store name, item type, and time remaining. You can approve or reject directly from the lock screen without opening the app, and the app will ask for notification permission only after your first successful ruling, never at launch. Quiet hours (22:00–07:00 local time) and your snooze preferences are respected before any notification is sent.
The mobile app now loads your last session and store instantly from local cache, so you reach your dashboard without waiting for a network round-trip. Switching between stores is preserved across restarts, and being offline no longer signs you out unexpectedly. Changes you make while offline are queued and sent automatically when your connection returns.
Profile, security, active sessions, notification preferences, and workspace invitations are now all found under your Account section — no more hunting across four different pages. Workspace-level settings (members, roles, billing, audit log, and more) have their own dedicated Settings group in the navigation. Store settings now contain only store-specific options, and a level switcher lets you jump between all three tiers from any settings page.
A new card on your Pulse page now invites you to download the Magistry mobile app, with a scannable QR code to take you straight to the App Store. The card disappears once you've installed the app or dismissed it, so it stays out of your way. Android support will follow when the Play Store build is available.
You can now tap action cards in Otto's mobile chat to approve or complete tasks directly from your phone. A Stop button lets you cancel a response mid-flight, and the Send button is disabled while Otto is answering so you won't accidentally send twice. Each action card shows its progress from Sent through Running to Done or Failed, and screen reader announcements keep things accessible.
The Otto screen now shows a clear banner when your device is offline, and displays helpful messages while content is loading or if something goes wrong. You'll always know what's happening instead of seeing a blank or broken screen.
The orange full-width readiness banner above your Ads overview has been replaced with a quiet single line showing your status and when it was last checked. Setup steps are still available on demand behind a toggle, and any billing alerts remain visible without needing to expand.
When a reply draft is blocked, withheld, or expired on a refund thread, Magistry can now pick it up again automatically — before the 6-hour first-reply target is missed rather than after. You can enable this per store and set how soon the retry kicks in (15 minutes up to the full horizon) under CS remedy settings.
When a Google Ads action cannot complete, you now see a plain-language explanation of what went wrong instead of a generic error. The message tells you whether the issue is something you can fix, such as a billing or policy problem, or whether Magistry is already handling it.
Actions in your reputation queue that could never be completed — because the review source has no reply box, because you already responded directly, or because they sat unanswered for 30 days — are now cleanly closed with a reason instead of staying stuck. You can see each outcome in your reputation history so nothing silently disappears.
When your store sends order acknowledgement emails, the promised reply time (24, 48, or 72 hours) is now chosen automatically based on what you have genuinely delivered to customers in the last 30 days. A time is only shown if you have kept that promise for at least 90% of customers who received one, so shoppers are never shown a commitment your team cannot meet. If there is not yet enough data to measure, no reply time is stated.
A new toggle on the Customer Service Feedback page lets you choose whether draft replies are checked against corrections using the originally stored email vector or a freshly generated one. When the toggle is off, the stored-vector results are still captured in the background so you can compare before switching. Turn it on when you want your team's corrections to match more consistently across similar threads.
When you have started connecting Google Analytics, Search Console, Ads, or Merchant Center but haven't finished choosing an account, Magistry now shows you exactly what to do next — either pick your account or visit Google directly — instead of offering a create option that would be rejected. API keys that have never been used after 24 hours are now flagged on the Connected Apps page with a Revoke link so you can clean them up.
When our ads health checks find a problem with one of your stores, the alert is now routed directly to your merchant card so you see it alongside any existing guidance rather than getting duplicate notifications. Campaigns flagged for targeting too many countries will now also surface a specific action card so you know exactly what to review. This means fewer repeated alerts and a clearer list of what actually needs your attention.
When none of the sites listed in your Google Search Console account matches your store, Magistry now explains this clearly instead of offering a site that belongs to a different store. If your connection has a problem that the site picker cannot fix — such as a wrong Google account or an expired grant — you will see what the issue is and how to resolve it.
A step-by-step migration wizard now guides you through your situation, stores, content, name, Google setup, and Magistry settings before starting a move. You can review the full plan — including any warnings to acknowledge — and type to confirm before anything begins. Progress is tracked live so you can see exactly where your migration stands.
When your connected Google Ads account has no active campaigns, you will now see a specific alert explaining what to do next: approve a pending Brain launch proposal under Marketing, create a campaign manually, or turn the Brain on. If billing is not yet set up, that step is highlighted first. The general 'account is silent' alert no longer fires alongside this more specific one.
When a refund can't settle because the customer already has their money back — through a won dispute or a payment processor refund — Magistry now shows you a card naming what actually happened for each order and linking directly to it in Shopify admin. Magistry does not retry, issue, or void these refunds; the card is there so you can close the loop with confidence.
When your connected Google Ads account has no conversion action set up yet, the tracking setup page now walks you through a Preview step so you can review what will be created before anything is written to your account. Only after you press Create does Magistry make any changes. This prevents unexpected conversion actions from appearing in your Google Ads account.
When your account is suspended, the banner now lists every area that has actually paused — such as Google Ads, tracking, agents, or mail collection — along with when each one last ran, so frozen figures elsewhere make sense. Tap 'What stopped' to expand the full list, or close it until tomorrow once you've read it.
When your account is suspended, the banner now lists each affected area — such as Google Ads, tracking, agents, and mail collection — along with when it last ran, so you know which figures are frozen. The banner collapses to a single line by default, expands with 'What stopped', and can be hidden until the next day using the close button.
When your store is suspended, the banner now clearly states that email collection has paused (including when email was last fetched) and that agents are not taking actions. It also shows the correct exit: buy a credit pack if you're on a paid plan, or upgrade your plan otherwise. If your subscription is cancelled but still active, you'll see a daily countdown to when your account moves to Free, along with a plain-language note that Free means agents will only propose actions rather than act.
When your store is suspended, the banner now clearly states that email collection has stopped (showing the last successful fetch) and that agents are not taking actions. It also shows the correct next step: buy a credit pack if you have an active paid plan, or upgrade your plan otherwise. If your subscription is cancelled but still active, you now see a daily countdown to when your account moves to Free, along with a clear note that on Free, agents only propose actions rather than act automatically.
When you cancel your subscription, you now keep full access to your current plan until the period you already paid for ends — you are not dropped to the free plan immediately. A cancellation confirmation email will remind you exactly when your access expires.
Magistry now checks your Pinterest credentials every day and alerts you when your connection has expired or been refused — so you know exactly when action is needed on your end. Platform or server-side problems are handled silently without raising a false alarm. When your connection is healthy or restored, any existing alert is automatically resolved.
You can now add and remove negative keywords on Pinterest ad groups directly from Magistry, and update ad group bids and bidding strategies without leaving the platform. Campaign and ad group creation is also supported, and Magistry now reads keyword lists, audiences, and conversion tags from Pinterest so your reporting stays accurate.
Your AI ad agent can now see your Pinterest audiences and their sizes, so it can make better campaign and targeting recommendations. It also checks whether your Pinterest conversion tags have ever fired, and warns you by name about any that haven't — so a silent tag no longer looks like working tracking.
Your Pinterest ad account's audiences now appear on the Audiences page, and the Pixel Health page shows whether your Pinterest tag is firing, partly firing, or missing. If your tag needs attention, you'll see a clear prompt to install the Pinterest tag or connect the Shopify Pinterest app. Pinterest negative keywords no longer incorrectly offer a broad match type, which Pinterest does not support.
When an active campaign's budget exceeds your envelope cap, the alert now tells you whether the budget was set by Magistry, by you in Google Ads, or by an unknown source. This makes it easier to understand why an overage happened and who needs to act. The AI assistant also uses this information when deciding how to handle headroom, so it won't treat a budget you set manually as something Magistry is responsible for.
You can now create a Pinterest campaign without leaving Magistry. Set your objective (awareness, consideration, conversions, or catalog sales), daily budget, target countries, and an optional bid — the campaign and first ad group are created in a paused state so you can review before going live.
You can now pause or resume a Pinterest campaign or any of its ad groups, update the campaign's daily budget cap, adjust an ad group's budget and bid, and add or remove negative keywords — all without leaving the ads drawer. You can also create a new ad group (it starts paused, so you can review it before it goes live). When Pinterest ad writes are not yet enabled for your store, every control shows you exactly why it is disabled.
You can now connect your LinkedIn company page and X account directly from the Social Accounts page using a real sign-in flow. Once connected, a per-store posting switch lets you turn LinkedIn and X posts on or off independently. Posting is off by default so you stay in control of when it goes live.
When Magistry detects that your store is spending on ads but Google cannot see any conversions, you will now see a dedicated alert card on your ad account showing the affected campaigns and their spend. Each alert includes a recommended next step to restore conversion measurement so your budget is not wasted on untracked activity.
You can now have Magistry automatically publish social posts to your LinkedIn company page and X account. Posting to each platform is controlled by a per-store switch, so you can enable only the channels you want. Images, hashtags, and character limits are all handled automatically.
Connecting your Pinterest Ads account now correctly requests catalog, board, and user account access alongside ad permissions, so the grant is no longer refused. If something still goes wrong during connection, the Ads settings page will now show a clear error message just like other ad platforms. Once connected successfully, your Pinterest ad data begins syncing right away.
Alerts now stack into cards by topic so you see one clear signal instead of dozens of duplicate rows. Repeat alerts within the same day no longer pile up, and alerts like tier demotion close automatically once the underlying issue is resolved. Every alert card now links directly to the relevant action or conversation.
When a product's cost hasn't been entered, profit, margins, and campaign return figures no longer treat the cost as zero and silently overstate performance. Affected rows now show cost unknown so you always know whether a number can be trusted. A cost-coverage indicator on the dashboard links directly to your costs page so you can see which products need a cost set.
Brain's Autopilot now has an optional weekly cap that limits how far your ad budgets can be raised within any rolling 7-day window. When enabled, a single step above last week's starting budget is the most any automatic increase can reach, while cuts, pauses, and manual operator changes are never held back. You can turn it on from the Autopilot limits section, and any time a raise is held you'll see a plain-English explanation in the decision log.
The Marketplaces page now surfaces bol.com ad spend, search terms, 28-day spend trends, promotion details, and category tree size alongside the existing channel summary. Returns can be opened to see their full details on click, and invoice, pick-list, and label PDFs are fetched on demand rather than missing. Rate-limit errors on the unpublished-offers report are now handled gracefully so the report is no longer lost.
The Marketplaces page now reads bol.com's own not-for-sale reasons for your offers and groups them by cause — missing attributes, stock issues, paused listings, and manufacturer address problems each get a preview and a one-click fix you can review before sending. Account-level blocks are shown as notices so you always know what is holding an offer back.
The season plan panel now displays the Brain's week-by-week budget plan, including the spending multiplier, campaign phase, week-on-week change, the Brain's reasoning, and which products are trending hot or cooling. Weeks the Brain hasn't planned yet are clearly marked as 'not planned yet'.
The Brain page now shows a read-only season plan panel covering the next 12 weeks. For each week you can see market demand signals, your own categories that are peaking at 15% or more above baseline, any promotions on your calendar, and best sellers within those peaking categories. Each signal shows its source and how recently it was read.
A new Returns page shows all open buyer return requests and open cancellation requests across your bol store in one place. From any order, you can now action a cancellation directly on the requested line item with the correct reason pre-filled. Returns and cancellation writes are controlled by a per-store setting that starts off, so nothing previously working is affected.
When one or more of your orders has passed its delivery promise, a warning card now appears on your dashboard so you can act immediately. For bol.com orders the promised date comes from bol's stated deliver-by date; for Shopify orders it is calculated from your store's own fulfilment history. Clicking the alert takes you straight to the affected orders.
Pinterest is now available to connect in the Ads section, so you can manage your Pinterest campaigns alongside your other ad platforms. TikTok will appear once it is ready to launch. No extra steps needed — just head to Ads and connect.
A new section on the Marketplaces page ranks the reasons your bol store may not be generating sales — including account status, listing visibility, late orders, buy box, pricing, stock, ratings, and billing. Each cause shows the numbers bol used to judge it, and per-service scores include bol's norm alongside what is pulling your score down.
A banner now appears in your dashboard when your AI credits are running low and suspension is approaching, giving you time to top up before anything stops. Once suspended, the banner also lists exactly which jobs have been paused so you know what is affected. Admins can see a full runway view for every store at a glance.
You can now create, edit, pause, and undo changes to your bol.com sponsored-products campaigns and ad groups directly in Magistry. Every change shows a preview before it applies, spend increases and archive actions require an explicit confirmation, and new items always start paused so nothing goes live by accident.
Magistry now automatically checks whether products in your feed are being submitted under a brand name that doesn't match their actual origin. Any mismatches appear as flags in the Google Merchant Center compliance console, so you can review and act before they cause disapprovals. You can also enable a hold setting to automatically keep flagged products out of your feed until resolved.
You can now download the official pick list and load-carrier labels (warehouse or transport) for any bol.com replenishment directly from the replenishments page. Both documents are provided as PDFs, exactly as bol.com generates them. No extra steps or configuration needed.
If your Google Merchant Center account is suspended and a full audit has been completed, Magistry now warns you starting on day 3 if no review request has been submitted — and reminds you again at day 7. The warning links directly to your recovery page and the review request form so you can act without delay.
You can now connect LinkedIn Ads via a standard sign-in flow and X Ads via your X account, then manage campaigns, budgets, and creative status directly from Magistry. LinkedIn and X ad accounts join the existing list of supported ad platforms in your Ads settings. StackAdapt support is in progress and will appear once write access is available.
When enabled, Magistry now builds a week-by-week budget plan shaped around your store's own historical demand, blending in broader search trends and similar stores. Each week's budget step stays within safe limits, and the plan highlights which products to push now and which to cool down. The feature is off by default — you can switch it on in your marketing settings.
Your Bol.com listings now surface promotions, ratings, price stars, buy box status, not-for-sale flags, and invoice requests — all updated daily. This gives you a clearer picture of listing health without leaving Magistry. The section appears automatically once your account is ready.
Your bol.com invoices now show a full line-by-line breakdown, so you can see exactly what each charge or credit covers. The invoices overview shows whether each invoice's specification adds up to its total, and a per-invoice page breaks costs down by line type and by order. Twelve months of line-type history is available at a glance.
You can now save your store-wide default GPSR declaration and override it per product directly from the Product Safety page. You can also declare or withdraw product recalls, and compliance counts update immediately after saving rather than waiting for the next weekly refresh.
You can now register a customer return and update a shipment's track-and-trace code for bol.com orders without leaving Magistry. Both actions are previewed before submission, deduplicated automatically if you click twice, and a full audit trail is recorded. The feature is behind an early-access gate, so you will see a preview notice until it is switched on for your store.
Lightspeed is now listed in Magistry's sub-processor documentation, keeping your compliance records accurate and up to date. Merchants connecting a Magento store also now use a cleaner tax-basis picker on the connection page.
Storefront Shield is now delivered through Shopify's theme app embed instead of a script tag, keeping it compatible with Shopify's upcoming platform changes. If the embed is not yet turned on in your theme, your dashboard will show a prompt and a direct link to activate it. Once enabled, protection works exactly as before with no changes needed on your end.
A new read-only content page shows every attribute Bol has on file for your products, including category, published status, and enrichment level. You can now check how your listings look on Bol without leaving Magistry.
You can now see, month by month, exactly what Magistry changed in your store, how those changes compare against untouched products, and the effect on each metric with an 80% confidence range. Months where an estimate isn't possible explain why, so you always know what the data is telling you.
Your SEO dashboard now includes Bing Webmaster data: daily crawl stats, submitted sitemaps, and per-URL index status sit alongside your existing Google Search Console coverage. Search Console also now separately tracks image, video, and news index totals rather than rolling them into a single figure.
Approval requests now arrive as notifications with Approve and Reject buttons. Tapping either opens the app to the item so you can act securely — approvals always go through biometric confirmation and rejections ask you to confirm once more, so nothing is decided from the lock screen. Every decision is now recorded with a full audit trail, including who acted and how.
You can now ship a bol order, buy a shipping label, cancel an order line, or handle a return directly from the order detail page. Before anything is sent, a preview shows the exact request so you can confirm before committing. Each action is recorded and audited so you always have a clear trail of what was submitted.
You can give your store a colleague a name of your choosing, and each week it writes a report in your chat summarising what changed, what is waiting for your decision, and any open alerts. The colleague also gets its own email address you can use to identify messages from it.
A new capability matrix view lets you see every supported API operation for each connected marketplace channel, along with its current status. Bol.com is fully detailed now, with Kaufland and Allegro shown as paused. The view is read-only for now.
You can now link your Sendcloud shipping account and Exact Online ERP to Magistry. Once connected, fulfilment parcel data flows in automatically and your ERP product costs are staged alongside matched catalogue variants for review. Sendcloud and Exact Online are listed as sub-processors in your data processing information.
Your obligations register now includes textile Extended Producer Responsibility entries for France and the Netherlands, calculated from your product categories and shipping origins. Each country where you sell as a distance seller gets its own line with the correct EPR class and status. EU, Latvia, and Hungary entries are evaluated but correctly show as not applicable where the scheme is not yet in force.
If you haven't created an account with Meta, Microsoft, Reddit, TikTok, Pinterest, or Snapchat — or with bol, Kaufland, or Allegro — you'll now see that platform's own sign-up page instead of a blank picker. Once you've signed up, a guided link brings you straight back to connect your new account to Magistry.
Magistry now collects packaging material and weight information for each product variant sold into EU markets, supporting your PPWR obligations. For Shopify stores, packaging details are read automatically from your metafields wherever you have defined them. Each variant is assessed for completeness — a valid entry requires both a material type and a weight in a recognised unit.
If your business migrates to a new Shopify shop while keeping the same Merchant Center account, Magistry now carries your existing offer IDs across to the new catalogue. This means your product-level Performance Max history is preserved, you avoid duplicate listings, and Google does not treat your entire catalogue as new inventory.
Your bol.com fulfilment-by-bol stock is now pulled in daily as a snapshot per product, so you can see exactly what bol holds for you. Active replenishments are tracked with their current state, carrier, and received quantities. Cover is judged against bol's own four-week demand forecast, giving you an early warning when stock is running short.
If your store's billing address is in France (including Guadeloupe, Martinique, and Réunion), your obligations register now shows the French e-invoice receipt requirement. The obligation appears with a status of not yet assessed, as confirming your designated certified platform requires a step you complete outside Magistry.
You can now connect your Outbrain Amplify account using your API credentials directly from the ads settings page, giving Magistry the ability to manage campaign status, daily budgets, and promoted-link status. Snapchat Ads is also available as a platform via OAuth once your workspace is configured with app credentials. Both platforms join the existing lineup so you can manage more of your advertising from one place.
You can now connect and manage up to twelve domains per account through the same single automation gate, instead of needing separate setups. This means less configuration overhead when running multiple storefronts or brands from one account. All domain-level automations continue to work exactly as before.
When reviewing a proposal, you now see an estimated effect band — such as a likely lift or neutral range — for each metric before you tap or click Approve. Estimates are drawn from similar proposals and are only shown when there is enough data to be meaningful. This makes it easier to set expectations before committing to a change.
Agencies can now upload their own logo to brand the Magistry console for managed clients, giving clients a consistent white-label experience. Agencies can also be suspended as a party, immediately blocking console access for that agency and its managed accounts. Client workspaces show a clear banner when they are operating under agency management.
When viewing a bol.com order, you can now see each order line with its unit price, line total, discount, commission, and fulfilment type (FBR or FBB), alongside any shipments and returns. The orders list also shows the marketplace channel ID and a status badge for quick identification. No changes can be made to the marketplace from this view — it is read-only.
Your bol seller data — including performance indicators, visits, buy-box share, sales forecast, top search terms, product rankings, and invoices — is now pulled in daily and shown on the Marketplaces page. The Brain also has access to these insights when answering questions about your channel performance.
The Integrations tab now uses a full-width layout with clearly grouped connection cards covering your store, Google, advertising platforms, tracking, social, inbox, API, sales channels, and more. You can connect every ad platform directly from this page, with the same OAuth and API-key flows available in Ads settings. Finding and managing your integrations in one place is now much faster.
When you clone a store, products and blog articles often keep the original brand's vendor name. You can now rename a vendor across all matching products in bulk, and update text in article titles, bodies, and summaries — both changes are safe to re-run and stop automatically if Shopify slows things down.
A new Why column on the product detail page now shows the source behind every price change, whether it came from a synced channel, a manual receipt, or an API update. You can also filter receipts by product to trace price history without digging through unrelated records. Prices that arrived via a sync no longer show as unknown.
The deep compliance audit now compares your Google Merchant Center shipping settings — delivery times, flat rates, and free-shipping thresholds — against your published shipping page, and cross-checks GTIN, MPN, and brand from your feed against your product pages. Returns-policy detection now works in Italian, German, French, Dutch, and Spanish, reducing false findings for non-English storefronts. A shipping page that cannot be read is flagged as unreadable; one that is silent on time or cost is marked unverified.
You can now link OpenAI Ads, Taboola, and Criteo directly from Ads Settings by entering your API keys — no OAuth flow required. Magistry validates each key before saving it, so you know the connection works before sync begins. Disconnecting a platform cleanly removes it from your active sync.
The mobile app now opens directly to the approval queue, so pending decisions are front and centre the moment you launch. The item with the nearest deadline is shown as a card inline, and the queue tab displays a live badge with your outstanding count. Your other tabs — Home, Fleet, Magistry, and More — are still one tap away.
You can now link your Pinterest account under Social → Accounts, and pick the board where Magistry publishes your pins. If you have multiple boards, you choose one during setup and your choice is remembered if you reconnect. Pins will not be published until a board is selected.
Setting up your theme's GitHub connection no longer requires a manual repository transfer step. Click Connect with GitHub on the Theme page, sign in to GitHub, and the repo is moved to your account automatically. The manual route remains available as a fallback if needed.
When a Facebook Page refuses to connect, Magistry now leads with the most common cause: the Page was left unticked during the Facebook login dialog and needs to be enabled under Business Integrations → Magistry → View and edit. The Page role requirement is still listed as a secondary cause. This makes it faster to resolve connection issues without contacting support.
Ad spend and conversion value figures on your performance dashboard are now displayed in your store's own currency rather than the Google Ads account currency. Historical data is converted automatically, so all your reports are consistent from day one. Days where no exchange rate is available are excluded until a rate becomes available.
After submitting a Google Merchant Center review request, the console now shows a confirmation card instead of the button — including who filed it, the expected response window (up to 7 business days), the estimated date, and Google's live programme state for your region. You will also receive an email when the request is submitted and again when the outcome is known.
When an AI assistant or connected app sends a reply on a ticket, it is now marked as machine-written rather than attributed to the agent who approved it. You can trace the reply back to the specific assistant action that proposed it, making it easier to audit and review automated responses.
You can now write store rules from Settings → Store rules to automatically deny or clamp field values across your catalogue. Integrations can also manage rules programmatically via the API. If you have manually kept a field value, that field is protected and returns a conflict rather than being overwritten.
Connected apps can now POST to /api/v1/simulate to run a full plan evaluation — store rules, policy checks, ownership, and ceilings — against live data inside a transaction that is always rolled back, so nothing changes. The response lets you see exactly what would happen before committing. Dry-run mode now also blocks all dispatching, including press actions, so it can no longer accidentally write to your store.
When a customer opens their support ticket via the link in your emails, each machine-written reply is now labelled clearly, and a plain-language notice in the customer's language explains that some replies may have been drafted by AI on behalf of your shop. Tickets handled entirely by your team show no AI notice. The hedge footer that appeared on every ticket regardless has been removed.
Orders placed on bol.com are now synced into Magistry automatically, keeping your marketplace sales in one place. Payment status, shipped quantities, and order totals are mapped correctly, and cancelled items are excluded from your revenue. Stores already carrying bol orders through their own shop will not see duplicates.
You can now set, update, and remove spending ceilings from Settings → Store. Apply a ceiling to a specific category, or leave the category blank to cap all categories at once. The section also shows the last ten times a request was refused because a ceiling was reached.
When syncing your bol.com offers, Magistry now fetches product titles directly from bol's catalogue so your imported products arrive with proper names instead of blank or placeholder titles. This applies to new products created during a sync for offers that don't already match a variant in your store. Existing products and variants are not affected.
When a connected app submits an approval request, it now shows up in your mobile queue alongside other items waiting for your attention — including the badge count and push notifications. You can approve or decline directly from your phone, the same way you handle other approval alerts.
Magistry now checks hourly whether campaign budgets, bidding settings, ad statuses, product prices, and theme files still match what it applied. If another tool or person has changed a value, you'll see an alert explaining what changed and who made the edit. From the alert drawer you can either keep the other change or restore Magistry's value.
Apps integrated with Magistry can now submit an approval request that pauses until you explicitly approve or reject it from your Alerts page. You stay in control — no action in your store happens without a person deciding, and requests that go unanswered expire automatically. Developers can track request status, list pending approvals, and withdraw requests via the API.
After a peak trading window closes, Magistry now surfaces a panel on your marketing dashboard listing actions that are due for review. You can choose to keep each action or revert it, and Magistry will apply your decision automatically. Actions not reviewed in time are lapsed so nothing is left silently in force.
When a prospect replies to a campaign email, their message now opens a conversation in your inbox instead of disappearing. Your AI drafts a response using everything it knows about the prospect, and a team member approves or edits it before anything is sent. No reply goes out automatically unless you have enabled autonomous replies.
You can now enter the Microsoft Ads conversion goal name that Magistry should upload purchases against, directly from Connections → Other ad platforms. The name must match your goal in Microsoft Advertising exactly, and purchases will only be sent once it is set. Leaving the field blank pauses the upload.
The ads onboarding wizard now lists all available advertising platforms and shows exactly where you are in the setup process for each one — connected, account chosen, synced, or ready to go. You can see at a glance which platforms still need attention before your campaigns can run.
Emails sent through outbound sequences now include a standards-compliant one-click unsubscribe link in the email header, so recipients can opt out instantly from their email client. Clicking unsubscribe suppresses the contact from further outbound emails, the same as an opt-out reply. For merchants outside the US, the postal address footer line is no longer required and will be hidden when left blank.
When migrating a Shopify store to a new location, your full order history — including buyer details, shipping, taxes, transactions, fulfilments, and discounts — is now carried across automatically. Orders are recreated exactly once with no duplicate risk, and no notification is sent to your customers during the process.
Kaufland and Allegro are now listed as sub-processors in Magistry's data processing documentation, alongside existing marketplaces like bol. They appear only when you connect a seller account, because order reads carry buyer personal data.
When a restore completes but some items failed, you can now click "Retry N failed" to requeue just those items without starting a full restore from scratch. The retry is tracked independently so you can see its progress separately from the original run.
Developers can now generate a sandbox API key directly from the Developers settings page without needing a connected store. The sandbox key provisions a pre-seeded demo store instantly, so you can explore and test API integrations right away. Sandbox keys are automatically scoped to demo data only and expire after a set period.
You can now link a WooCommerce store by entering your REST API credentials on the new WooCommerce connection page in Settings. Magistry verifies the keys immediately and saves them securely. Product and order sync will follow in an upcoming update.
When setting up a new store, you can now pick a complete backup from any of your other stores and restore it here — no more dead end. A confirmation step asks you to type the store name before anything moves, and the restore list tracks direction, progress, counts, and errors in real time while the restore is running.
You can now hand over a Google Ads or Merchant Center account that Magistry set up directly into your own Google ownership. Magistry confirms you are already an accepted admin before making any changes, then removes its own link only after you explicitly confirm. The final account status is read back from Google so you always see the real outcome.
You can now check status, start, and cancel a provisioning operation programmatically via the REST API at /api/v1/provisioning. Starting a provision is idempotent — if one is already open, the response tells you so rather than creating a duplicate. All operations are covered by the OpenAPI contract.
Your marketing mix model now fits each ad platform separately, so the Brain and Campaign Agent can see which channels are actually driving revenue rather than a single blended view. Platforms with fewer than 30 days of spend are skipped with an explanation so you know why they are excluded. Results are only used when the model's numbers reliably reconstruct your observed revenue, keeping recommendations trustworthy.
If Magistry provisions a Google tag for your store but no signal from it is ever received after a week of activity, you will now see a health card explaining what to do. The card tells you either to finish the installation (enable the app embed and add the purchase snippet in Customer events) or explains if Magistry has paused the tag because it was never detected. This replaces a silent internal alert that was never visible to you.
Instead of asking you to press Publish yourself, Magistry can now publish your library theme directly — including safety checks to ensure your product pages will load before anything goes live. If Shopify declines the action for any reason, you'll see the exact message in plain language.
Amazon Ads, Taboola, Criteo, AppLovin, and OpenAI Ads are now recognised across the dashboard, with platform icons and correct labels shown wherever your active platforms appear. Platforms that cannot yet sync are shown with a disabled toggle and a clear reason rather than a control that does nothing. The sub-processor list in the compliance section has been updated to name every host that may receive your campaign data.
You can now connect your AppLovin AXON account to manage campaign status, daily budgets, and bidding strategy directly from Magistry. If you have only a management key and no reporting key, Magistry will tell you clearly that performance data is unavailable rather than failing silently. Budget and goal changes are validated against your campaign's current setup before being sent, so mismatched configurations are caught and explained before anything is written.
You can now connect your Criteo Marketing Solutions account to manage campaigns and ad sets, and pull performance statistics directly in Magistry. Because Criteo activates ad sets rather than campaigns, pausing and resuming controls work at the ad-set level. Statistics reporting is included so your Criteo data appears alongside your other channels.
You can now manage your Taboola (Backstage) native discovery campaigns, items, and daily performance reports directly in Magistry. Campaign status reflects Taboola's full approval state, so a campaign pending review shows correctly rather than appearing live. Budget changes are verified against Taboola's own response before being recorded as applied.
You can now connect your Amazon Sponsored Products campaigns to Magistry, giving you visibility and control over your keyword-based Amazon ads alongside your other channels. Campaign budgets are protected — Magistry reads your budget type before making any changes and will refuse to touch a lifetime-budget campaign in a way that could convert it to a daily cap. Regional API endpoints can be configured per store to ensure your account data is always read correctly.
You can now connect your OpenAI Ads account to Magistry to sync campaigns, ad groups, ads, and performance insights. Budget and bid figures are converted automatically so the numbers you see match what OpenAI reports. Declared capabilities for Meta, Reddit, Microsoft, TikTok, and Pinterest have also been corrected, so the actions available to you reflect what each platform can actually do.
You can now enter a carrier tracking number and internal notes on any Shopify return directly from the Returns page. The tracking information is automatically pushed back to Shopify so your customer sees the correct return shipment status, and each return shows whether the sync is queued, complete, or failed.
When product research is blocked by a setting you haven't turned on yet, the candidates page now shows a clear message and links directly to the switch that controls it. You'll also see a specific notice if there are no seed products to work from, or if research has never been run for your store. No more blank page with no explanation.
The candidates page now shows a breakdown of why sourcing candidates were rejected, in terms you can act on — such as missing store type, no price or image, or sourcing feeds that are not currently reporting. Where feeds are silent because a credential is missing, that is also called out so you know what to restore.
When you press the same ad decision multiple times across separate refusals, the card now tracks the full sequence — showing a press count from the second press onward. Previously each press appeared as if it were the first, hiding the pattern of repeated holds on a single store.
Workspace owners can now enforce SSO for all members, so password-based sessions are rejected at the door. Before the setting takes effect, Magistry confirms that your domain already has an identity provider configured and that you are currently signed in via SSO — so you cannot accidentally lock everyone out. Turning the requirement off is always permitted.
Each step in the launch wizard now tells you whether it's your turn or Magistry's, how long the step normally takes, and what done looks like. For example, you'll now see that Merchant Center verification can take a few days, and which settings you're filling in feed directly into your product feed. Nothing to do at an automated step is shown explicitly — so you're never left wondering if you missed a prompt.
Workspace owners can now restrict dashboard access to specific IP addresses or CIDR ranges. Adding your first range is safely guarded — Magistry checks that your current IP falls within it before saving, so you cannot accidentally lock yourself out. Removing a range is always permitted, giving you a way out of any configuration.
You can now enable SSO-required sign-in and manage SCIM tokens directly from the Identity section of your workspace admin settings. Revoking a token or toggling SSO enforcement gives you a clear explanation if the action is not permitted, rather than a silent disabled button. New tokens are shown in plain text once at creation time.
You can now restore a backup directly onto the connected Shopify store, not just onto a different destination. Before anything is overwritten, Magistry automatically takes a fresh backup of your live catalogue and keeps it safe from pruning, so you have a guaranteed way back. A two-step confirmation — including typing your store name — is required before the restore can begin.
Agency owners can now invite team members who only need to see invoices and billing details, or only need to receive report notifications, without granting them access to the full client book. Each seat is independent: a billing seat holder sees what your agency is charged and nothing more, and a notify seat holder receives agency report emails without any console access.
When accepting an agency connection, you are now asked whether the agency acts as a data processor, joint controller, or sole controller for your store's data. Existing connections show "roles not recorded" until you confirm. No answer is pre-selected, because this is a legal determination only you can make.
The AI spend summary on /admin/costs previously under-reported costs whenever key-level usage data was missing for part of a month — the missing days were silently dropped rather than estimated from your usage meter. Costs are now attributed day by day, gaps are filled from your own meter, and the panel shows how much was reconstructed and how many days have full coverage so you can see any data quality issue at a glance.
Workspace owners can now enforce that every session in their workspace must come through your connected identity provider — password-based logins are blocked at the door. Turning the requirement on is gated: your workspace must already have a verified SSO domain configured, and your own current session must be an SSO session, so you can never lock yourself out. Turning it off is always permitted.
Every stage in the launch wizard now tells you whether you need to do something or whether Magistry is working on your behalf, how long that stage normally takes, and what done looks like. Steps that run automatically say so explicitly, so you are never left waiting for a prompt that will not come.
Workspace owners can now define an IP allowlist so that only requests from approved addresses or ranges can reach the dashboard. Ranges are validated against your current IP before saving, so you cannot accidentally lock yourself out. Removing a range is always permitted, giving you a way back if your network changes.
Workspaces can now submit an application for agency manager access directly from settings, instead of hitting a dead-end 'Coming soon' screen. If an application is declined, the reason is shown clearly and you can apply again. Approval unlocks the agency experience immediately while full provisioning follows the established sign-off process.
Agency owners can now give a team member access to invoices and billing without handing them the full client book, or add a member who receives report notifications without any store access at all. These two new seat types sit alongside existing store grants, and a seat holder cannot also hold a store grant — so access stays exactly as wide as intended with no extra steps.
When you connect an agency to your store, you are now asked to record whether your store, the agency, or both jointly determine how customer data is processed. This answer is required at the moment you accept the connection, with no option pre-selected, so the record reflects your decision rather than a default. Links made before this change show "roles not recorded" until you update them.
The catalogue stage now lists every draft heading to your catalogue, showing whether you or Magistry chose it. You can remove any of Magistry's automatic picks before launch, and you can turn off automatic selection entirely if you prefer to choose every product yourself.
When preparing a product launch, Magistry now requests exactly as many products as needed to reach your catalogue target in a single review session. Instead of approving batch after batch to fill your queue, you can work through all candidates at once and move straight to launch.
The catalogue stage now shows every draft already queued for your launch, with a label indicating whether you or Magistry selected it. You can remove any of Magistry's automatic picks in one click, and a toggle lets you turn off automatic selection entirely if you prefer to choose every product yourself.
When selecting products for a new launch, the review queue now asks for exactly as many products as needed to reach your target in one go — no more returning to approve the same queue multiple times. The system enforces a sensible ceiling so the queue stays manageable, and an empty result is shown clearly rather than treated as an error.
When reviewing import candidates, the product image now fills the card so you can judge the garment itself rather than a cropped thumbnail. Each candidate shows why it was ranked — either a sales figure from competitor data or its position in the source shop's best-seller list — so approving the top products is a decision you can stand behind. You can now approve in bulk, and the page shows how many were created immediately alongside how many are still being added to your store.
When you review candidate products during a launch, items that the sourcing shops themselves promote as best-sellers, trending, or most-loved are now ranked ahead of arbitrarily ordered results. This means the products most likely to perform well appear first, so you spend less time scrolling to find the strongest candidates.
When Magistry declares a known issue, the notice now shows on the specific page affected rather than only as a site-wide banner. A current notice about Theme Studio's GitHub connection is visible on the Theme and Launch pages, where you would first run into it.
When you review products sourced from competitor shops during a launch, cards are now ranked by each shop's own publicly-stated best-sellers so that the strongest candidates appear first. Previously, products were shown in arbitrary scrape order, requiring you to read through all staged cards to find the ones most likely to perform. Shops that do not publish a best-sellers collection are still fully included — they simply keep their current ordering.
If you're on the Theme Studio page or the launch wizard and a GitHub repository connection cannot be completed, a notice now appears right there explaining what is affected. Existing connected repositories continue to deploy without interruption. New connections will not complete until the issue is resolved.
The platform status page now has a dedicated Storefront component, so problems with Theme Studio and storefront publishing are reported under the right name instead of being grouped with catalogue updates. Known issues about the GitHub connection will now appear exactly where you would expect to find them.
When Magistry declares a known issue, a banner now appears directly on the page affected — so if a feature isn't working, you see the notice right there instead of having to check a separate status page. The banner is shown only where it is relevant and disappears automatically once the issue is resolved.
When transferring your theme repository to your own GitHub account, you now receive direct instructions and a link straight to the right settings page — no waiting for an email that was never sent. The handover card shows exactly two steps, both yours to take, so nothing is left ambiguous.
When something breaks and you report it, the status panel can now reflect a declared incident in the same place the automated health checks appear — one surface, never two that disagree. If your report is linked to a known issue, your feedback thread will show that it is recognised, in the operator's own words rather than a generic message.
You can now prove ownership of your company's email domain by publishing a DNS TXT record at _magistry.<yourdomain>, unlocking automatic workspace membership for colleagues who sign up with that address. Public email providers like Gmail are refused clearly rather than silently failing. The workspace admin page shows exactly what DNS answered so you can tell at a glance whether your record is in place.
Merchants connected to an agency can now see a Recent access list in their store settings showing which agency operators have entered their store and on what date. Agency users see the same information from their client list, including a clear notice when nobody has opened a store yet. Dormancy is only flagged after the ledger has had enough time to be meaningful, so a brand-new connection is never incorrectly marked inactive.
Your client book now groups multiple storefronts belonging to the same customer into a single company row, so you see one client rather than several unrelated lines. Priority is set by the worst-performing store in the group, keeping problem accounts visible at the top. Enabling a manager account now presents versioned terms you must accept before the conversion completes, and every acceptance is permanently recorded alongside the document version you agreed to.
The platform status panel in the header has been redesigned to match the notifications panel: a compact tab switcher with Status and What's new, a persistent outage banner pinned above whichever tab you're on, and a footer link to the full page when you need more detail. The panel always opens on Status if anything is not operational, so you never miss an incident because you last left it on release notes. Releases on the same day now appear in the order they were written rather than an arbitrary internal order.
The agency workspace page previously showed two sets of revenue numbers and store grids stacked together with no label telling you whose was whose. Now the page leads with your client book — each client's account number, their stores, and their ROAS — and your own stores sit under a clearly labelled section beneath. ROAS is shown as a dash rather than zero or infinity when a client ran no ads, and a client company name appears on a row only when that company owns more than one store.
A new status page at /dashboard/status shows whether each part of Magistry is operating normally, degraded, or experiencing an outage — updated automatically from the jobs the platform already runs. A coloured dot in the header bar appears whenever something is degraded or down, so you can tell immediately whether a problem is on our side or yours. If the status system itself stops updating, every component shows as unknown rather than falsely green.
When an agency requests access to a client's store, the client must share a short code alongside their account number — without it, the request is refused before it reaches them. This closes the door on unsolicited access requests from unknown agencies. Codes are on by default and can be rotated at any time from the client's workspace settings.
Every store and workspace now has a unique 10-digit account number you can read on the Settings page and quote to an agency or support team over the phone. A separate 6-character request code controls who may ask to link to your account — you can rotate it at any time, or turn the channel off entirely from Settings. Switching the channel back on after it was off issues a fresh code automatically.
Magistry now uses a single shell for both your store and your workspace, so the header, menu, and layout look and behave the same everywhere you go. Agency account holders will find their clients, reports, activity, and billing pool directly inside Workspace rather than in a separate area. The workspace overview now shows managed client stores at the top, so the portfolio no longer appears empty when stores are linked.
The in-app changelog now shows real release notes written for merchants, not raw commit messages. The roadmap carries measured, dated evidence for every claim rather than freeform descriptions. An unread dot appears in the navigation when new entries have been published since you last visited, and the dot clears the moment you open the page.
The Links page now shows every linked product group with images, lets you unlink individual products or entire groups, and lets you turn sync on or off per link rather than all at once. It also clearly explains what a link always shares (what products have learned from each other) versus what only activates when you enable sync. The page is now reachable from the Catalog toolbar, beside Suppliers and Export.
The photo matching that connects the same product across stores now recognises re-encoded or lightly cropped shots as the same image, so genuine matches are no longer missed. The review screen now shows each product at a size where you can actually tell garments apart, and the confidence blurb explains what the score means in plain terms rather than how it is calculated.
Cross-store image linking is now something each workspace explicitly enables, rather than being on by default for any multi-store account. If you have asked to use this feature, your workspace remains active; all other workspaces are off until you choose to turn it on. You can opt in using the Scan workspace button in your links settings.
System status measures itself from the work the platform does, so it sees a sync that fails and not a button that does nothing. Those now get said out loud: a known issue carries what is broken, what you would notice and what is being done about it, at the top of the status panel rather than behind a colour. Report something that turns out to be one of them and your own thread says so, in the same words.
When something looks wrong, the question is whether it is your setup or our platform, and nothing in Magistry answered it. System status now shows each part of the platform, what it covers and whether it is working, with an incident timeline when it is not. It measures itself from the work it does rather than from anything anybody remembers to update, so it is current whether or not we are looking.
One button beside your notifications carries both: the state of the platform and the newest releases. A coloured marker means something is degraded; a quiet one means there is something new to read. It was reachable only from the account menu before, which is a page nobody finds.
Connect a Reddit ad account and the Brain can pause, resume and re-budget campaigns there on the same rails it already uses for Google — the same approval model, the same spend ceilings, the same decision log. Reporting reads back from Reddit rather than from what we asked for, so a change that did not take shows as a change that did not take.
Linked products could be created and never listed. The page now shows every link, what each one shares, a switch per shared field, and an Unlink on both the group and the single product. Sharing cost data is separate from mirroring a price or an archive, and each is its own control.
It was mounted on one of five app shells, so Launch, Workspace, Agency and the welcome flow had no way to report anything at all. Every surface with a signed-in reader now carries it, and notes are filed against the page you were standing on.
The agent's credit spend now renders on its own page beside its mode, because either number alone reads as fine: a spend is unremarkable on a desk that answers customers, and "Dry run" is unremarkable until you see what it costs. The four modes are named individually instead of collapsing into "Approval".
A new Links chip appears between Export and Suppliers on the Catalog toolbar, so you can check whether a partner store carries the same products without leaving the page you are already on. The chip is hidden when your workspace has only one connected store.
Products across your connected stores are now scanned weekly for matches using shared SKUs, supplier images, and product similarity — so you no longer need to trigger linking manually. A review queue shows you candidates with the evidence behind each suggestion before anything is confirmed. Confirming a link shares data between stores but never moves a price.
The ticket link panel now shows whether a customer has viewed, replied through, or never received their portal link — so agents know at a glance whether a reply has been seen. If a link was forwarded to the wrong person, agents can now revoke it; the customer automatically gets a fresh working link on the next send while the old one stops working immediately. Previously there was no way to cut a portal link from inside the dashboard at all.
Customer-facing replies sent from your helpdesk now include a masthead, a preheader, an author block with the signer's name and photo, and a ticket reference number — matching what customers expect from a professional support email. If you haven't uploaded a photo for your support agent, their initials appear on a branded monogram instead, so every reply looks intentional from day one. Upload your agent's photo under CS → Settings → Ticket page to personalise the author block.
The support ticket portal was missing a favicon, so it appeared as a blank tab icon next to your shop's branded tabs — this is now fixed with a full set of icons including mobile home-screen support. The portal also gains the same security headers the main dashboard already had, including stricter cross-origin resource policies and HSTS across all subdomains.
The Tracking page now shows every enrichment signal Magistry sends to Google, with a separate indicator for whether you have enabled it and whether your store actually holds that data. Signals exclusive to Magistry — including cart contents, customer value band, and profit instead of revenue — show the specific reason no tag vendor can match them, so you can verify the claim yourself. Coverage figures only appear when they have been measured.
Automated reply emails no longer include a disclosure stating they were written by an AI assistant. The link to your support ticket is now a clearly styled button that renders correctly in Outlook and on mobile screens, replacing a small grey URL that was easy to miss. All three email sending routes now include the ticket button.
Magistry now sends Google Ads a customer value signal — LOW, MEDIUM, or HIGH — based on each buyer's real spend history with your store, so your campaigns can automatically bid more for the customers worth having. The signal is ranked within your store, not against an absolute threshold, so it works correctly even for newer stores. When there is not enough data to be confident, no signal is sent rather than a wrong one.
Google conversion events now include the buyer's browser type and IP address alongside the hashed contact details already being sent. Google uses these signals to match events to clicks or users it recognises, and drops events it cannot join — so more of your conversions will now be attributed correctly. Device details are withheld whenever a customer has explicitly denied consent, the same gate that protects all other personal data.
The Brain, Website, and Social sections now display a "Beta" badge in the navigation, so you can easily identify which features are in early access. These features also appear under Settings > Beta features alongside other Beta areas.
A new setting on your conversion goals page lets you decide whether Magistry reports refunds back to Google after a purchase is tracked. Off by default, so your conversion count stays comparable with tools like TrackBee that never send refund adjustments. Turning it on gives Google a net-of-returns figure, which can steer bidding away from buyers who return — but will show a lower number than competitors tracking gross revenue.
Your integrations and AI agents can now discover Magistry's API capabilities automatically via the standard /.well-known/mcp.json address. The versioning guarantees previously buried behind a broken link are now embedded directly and machine-readably in the contract document, so your CI can assert on the notice period as a number without parsing prose.
Magistry's REST API now has a generated TypeScript SDK with typed arguments for every operation, an llms.txt file so AI agents can discover the API, and a server.json MCP registry manifest. All three are derived from the same OpenAPI document that the API already serves, so they stay in sync automatically.
The Brain, the ads specialist, the theme engine, the storefront agent and the customer-service desk each carried their own mode, and they disagreed on live stores — one desk was set to autopilot in one place and to dry run in another. There is one control now; each agent's page says what it inherits and links to the setting that changes it.
You can now add your own logo to the support portal your customers use to track and reply to tickets. Upload an image in the portal branding settings and it will appear in the masthead straight away. A preview of how the portal looks to customers is available on the same page.
A new control on the ticket page lets customers flag their request as urgent. On stores with a staffed inbox the thread is held for a person; on stores handled automatically the request is prioritised rather than paused, so it still gets a reply. The button's confirmation survives a page refresh, so the flag is never silently lost.
When a ticket is closed, customers now see a satisfaction rating panel at the bottom of the support page they were already linked to — no second email required. The rating appears only after a ticket is closed, and a submitted score is shown in place of the form on return visits. The follow-up rating email is no longer sent when the portal page can collect the rating instead.
When Magistry reads your policies and proposes knowledge rules, any rule that needs information from you — like a return address or store-specific detail — now shows an editor so you can fill it in before the rule goes live. Rules are grouped by how much attention they need, conflicts with your published policy are called out separately, and a search bar reads your own wording, not just our paraphrase. Approving in bulk no longer pre-accepts rules that flagged a genuine disagreement with your store's policy.
When your support team has put a refund or compensation offer on the table, you can now see the amount and its deadline directly on your ticket page — no need to dig through your inbox. Two reply shortcuts let you respond in your own words, and your reply is what confirms your decision.
When a customer writes in about an order, agents now see a panel with the relevant order details — including the carrier handover, delivery status, and any refund settlement — pulled directly from what was already sent to that customer. The panel only appears when the order can be matched with confidence, so agents never see a guess. Product photos from the order are shown inline.
When a customer opens their support ticket in the portal, any photos or files attached to the conversation are now visible to them — and they can attach a file of their own when replying. Previously, a filename was shown beside a message but the file itself could not be viewed.
Cleaning up outdated offers from your feed now runs with higher concurrency instead of one at a time, so large backlogs that previously took over an hour of wall-clock time finish in a fraction of that. Quota limits are still fully respected, and the count of removed versus not-found offers is reported accurately as before.
When a customer opens their support ticket link, they now see who owes the next reply and how long ago that happened — rather than a raw message count. The message body shows readable text by default, with the original email available on demand. The page also now discloses when replies may be drafted by an automated assistant, as required.
You can now connect your Reddit Ads account and manage campaigns alongside your other ad channels. Conversion events are tracked through Reddit's Conversions API, and the AI assistant can plan and dispatch Reddit ad actions just as it does for other platforms.
New stores now receive a personal offer to book a one-on-one onboarding call directly inside the welcome email. A named onboarding specialist reaches out so you can get set up quickly and with support.
You can now reach the Magistry team by email — and book a call when a scheduling link is available — without leaving the dashboard. The contact card appears on Settings → General and the Help & guides page.
When Magistry checks your support playbook against your published pages, it now runs a second verification pass before marking anything as a conflict. Conflicts that were really just your playbook choosing one valid option from several your site already offers are quietly downgraded instead of blocking your rules. Real contradictions — like a page showing a phone number while your playbook says phone support is unavailable — are still caught and flagged.
Filing applied the folder's tag and left the conversation in the inbox, so a filed thread sat in two places. Tags label and folders file, and the two are now different things. Routing rules can file automatically, they never overrule a folder a person chose, and every rule application is recorded with its outcome.
New stores now receive a dedicated customer service portal hostname automatically, without anyone needing to run a manual setup step. The hourly job mints, registers, and verifies the host, and the portal link appears in the dashboard once the host is confirmed live.
Every outbound reply email now includes a personalised link so the customer can view their ticket and conversation history directly, in their own language. The link is only included once the store's support portal has been verified as reachable, so customers will never land on a broken page.
When your playbook is compared against your published store pages, claims are now checked in smaller groups so contradictions that were previously missed now surface correctly. The panel also no longer claims to catch every disagreement — it tells you what it actually does, so you know to treat the results as a strong signal rather than a complete audit.
You can now upload your own support playbook — a Word file, PDF, or web page — directly from the CS Knowledge page or the CS setup wizard. Magistry reads your document, sorts each section into the right part of your agent's training, and flags anything that conflicts with your published store policies for you to review before it goes live.
A dedicated customer portal lets shoppers open their support ticket by following a link in any email you send them, read the full conversation, and post a reply — all without logging in. Each store's portal lives on its own permanent address that won't break if you ever rename your store. Email bodies are displayed safely, with no risk of one store's content appearing under another's address.
Your customer opens one page on your own subdomain that holds the whole conversation, the photographs they sent, a reply box and a rating. It is branded per store, it works without an account, and a ticket with nothing in it is refused rather than shown as an empty conversation.
The AI assistant you use across the dashboard, your merchant emails, and the mobile app is now called Magistry everywhere you see it. No features have changed — only the name on the screen.
The AI that drafts customer service replies was learning from automated form letters, customers' quoted text, and emails mislabelled as agent-written — none of which reflected how your desk actually writes. It now studies only the messages your team composed, with quoted conversations and HTML template wrappers removed before any comparison. Draft tone and phrasing should more closely match the voice your customers already recognise.
When the brain's conflict guard is holding a campaign slot, you can now see exactly which move is blocking, how long it has been held, when it will release at the latest, and how many proposals have been turned away during that window. Landing-page quality signals are now grouped by destination URL, so the brain can tell the difference between pages that need work and pages that are already clean.
When a cost spreadsheet can't be matched to your catalogue, the import assistant now explains exactly which identifier and cost columns it tried, what it was able to read from your catalogue, and which other columns in your sheet could work instead — so you have a real next step rather than a dead end. If the file genuinely has no other way to name a product, it tells you that plainly too.
A setting that read like a preference was deciding whether a customer got an answer at all: mail outside the configured language was escalated to a person, and on a desk with nobody behind it that meant silence. The customer's own language now decides the reply, whatever the store's markets are.
The examples the drafter imitates are now taken from what your team really sent, with the quoted history and the mail template cut away first, and per store and per language rather than per fleet. The gap between the two is measured every week, so a desk drifting away from its own register shows up as a number.
Each support thread now has a unique link you can share with a customer, letting them read the full conversation and send a reply that arrives in your inbox exactly like an inbound message. A customer reply will also automatically reopen a closed ticket so nothing falls through the cracks.
When a connected platform has been consistently refusing requests, your integrations page now shows a warning so you know something needs attention — instead of silently failing in the background. Stores whose Shopify subscription is inactive are automatically paused to stop wasting credits, and resume on their own once Shopify confirms the shop is available again.
Before you leave for Facebook, the connection flow now tells you exactly what role you need on each Page and where to set it in Business Suite. The Accounts page shows what each connected account can do rather than internal status messages, and Instagram accounts are shown under their parent Page so the relationship is clear. Pages you cannot connect yet are grouped into a collapsible summary with the one step that fixes them.
When Otto uses tools to look something up before answering, it now sends a short orienting line first so you know it is working. Previously the screen could stay blank for over ten seconds while Otto gathered information.
You can now ask Otto to explain the Brain's current strategy, standing instructions, schedule, why it held off last cycle, and what is queued next. You can also tell Otto to add, remove, or leave a one-off note for the Brain, and those instructions take effect on every future cycle. Changes are logged so there is always a record of who asked.
Workspace owners can now enable a shared credit pool so every store in the workspace draws from a single balance instead of each store managing its own. Turn it on from Workspace Settings → Billing, and each store's billing page will show its usage against the shared pool total. New stores you add later join automatically — no extra setup needed.
Workspace owners can now enable a shared credit pool so every store in the workspace draws from a single balance instead of each store managing its own. New stores added later join automatically — no manual setup needed. Turn it on from Settings in your workspace, where you'll also see a full breakdown of how credits are being used across your stores.
When a teammate with view-only access signs in, a one-time banner now tells them upfront that they cannot make changes — before they run into a confusing blocked action. The banner is dismissible and only appears once per browser, so it informs without becoming background noise.
Workspace owners who have been granted access to the Store Launch Engine can now use it directly from their workspace — no special staff permissions required. If your workspace has been opted in, the Add Store entry point will appear in your workspace settings.
Your subscribed webhooks will now receive an `order.refunded` event whenever a refund is recorded on an order, whether it was processed by Magistry or applied directly in your Shopify admin. The event is delivered exactly once per refund and includes a `settled` field that tells you whether the payment gateway had settled the refund at the moment it was first seen.
The checks Magistry runs before answering your first chat message now happen at the same time instead of one after another. You should notice a shorter wait between sending a message and seeing Otto begin to reply.
The Otto mode you pick — fast, smart, or deep — is now remembered for each store separately, so switching to deep on one store won't quietly follow you to another. Deep mode now uses extended thinking for more thorough answers, and each mode has a firm answer-length limit so a long report can no longer be cut off mid-sentence.
Each API response includes an `x-request-id` header, and that same ID is recorded against the matching audit row. You can send your own ID in the request header so both sides share it without translation. The ID also appears in the response body of the diagnose endpoint, because its output is pasted into support conversations where a header is not visible.
When a customer contacts you about a refund, you can now look up that specific order and amount in the agent activity export — not just that a refund happened somewhere that day. The export is available as JSON or CSV from the audit page, and every export is itself recorded so you have a complete picture of who reviewed what.
You can now rename any AI conversation from its row menu, with the name editing in place so you keep your place in the list. Pasting multi-line text collapses it into a single name rather than cutting off at the first line. If a name cannot be saved you are told immediately and the field stays open so you can correct it.
Otto now offers three distinct modes — Fast, Smart, and Deep — each with genuinely different reasoning depth and time budgets. Fast gets you a quick answer with fewer tool rounds; Deep works harder and longer on complex questions. A mode picker in the chat input lets you choose before you send.
If you create a webhook subscription whose event-type filters only match types that have no active producer, the API now returns 201 with a warning listing the dormant types by name. You can still subscribe ahead of a producer shipping, but you will no longer receive silent confirmation that implies events are on their way.
AI conversations now get a meaningful name based on your first exchange instead of showing your raw question as the title. The name is set once and never overwrites anything you have chosen yourself. If a good name cannot be generated, the conversation keeps what it had.
When creating an API key you now choose an expiry — defaulting to 90 days — so keys no longer live forever by accident; 'Never' remains available for integrations that genuinely need it. Rotating a key now keeps the old key working for seven days alongside the new one, so you can redeploy at your own pace without an outage. The successor inherits the same scopes and settings exactly, and the rotation is blocked if the key is already revoked or mid-rotation.
You can now set an expiry date on headless API keys, restrict them to a list of allowed IP addresses or CIDR ranges, and rotate them with an overlap window so your integration stays live while you deploy a replacement. When a valid key is rejected, the error now tells you whether it expired or was blocked by an address rule — without leaking that information to unknown callers.
You can now copy any answer Otto gives you, ask the same question again if the answer missed the mark, and tell Otto whether a response was useful or not. Thumbs down opens an optional box where you can say what went wrong — your feedback goes somewhere it is read, not into a void.
When Otto produces a report, table, chart, or plan, it can now open it beside your conversation as an artifact rather than leaving it buried in the thread. Artifacts stick around after the chat ends, every revision is saved with a note explaining what changed, and you can copy or download the whole thing as Markdown at any time.
When you edit a grant — removing a store or dropping a scope — apps that connect via the Magistry MCP endpoint can now detect the change immediately from a version token in each tools response, and re-read only when something actually differs. Pagination is also live for credentials with read and write access, so every available tool is reliably returned across multiple pages.
When you ask Otto about your data, it can now respond with a chart — line, bar, area, pie, donut, scatter, and more — chosen to match what the data is showing. Charts include a title, currency-aware formatting, and a data table so you can always see the numbers behind the visual. Charts are saved to the conversation and will still be there when you come back.
AI clients like Claude and Cursor can now read each MCP tool's risk level before deciding whether to ask for confirmation. Write operations — including refunds and price changes — are always flagged as potentially destructive, so a client will prompt before acting rather than treating them the same as a read. Repeated calls to write tools are also marked as non-idempotent, meaning a second refund call is correctly treated as a second refund.
A new "Test connection" button on the Developers page runs four checks in the order they actually block a connection — live grant, exposed surface, credential validity, and recent activity — and tells you whose problem each step is. Any step that could not be evaluated is shown as unreadable rather than passed, so a green result means the connection genuinely works.
The importer now resolves your columns itself where the file can settle them, matches on the Shopify product id when the sheet carries one, and reads a cost per variant instead of broadcasting the first row's price across every colour. A sheet priced in a currency your shop does not sell in is recorded and held back rather than written to Shopify as if it were the shop's own.
An account Google has switched off is not a feed-quality problem, and it used to be graded with the same report as a healthy one. The compliance page now leads with the account's standing and a day counter, sequences the steps that cannot be reordered, and says plainly which requests Magistry can make for you and which only you can press.
Storefront presets are a lever the Brain can propose like any other: it snapshots the exact files first, the edit carries a revert time taken from your own promotion calendar, and a timer performs the revert. One change per store per day, so a promotion window yields a reading rather than a pile-up.
A 200 from Google means the request parsed, not that the conversion was accepted — acceptance, the per-event errors and the match rate arrive later. Every upload now leaves a receipt, an hourly sweep settles it against Google's own answer, and the tracking doctor reports rejected, unsettled and accepted-but-unmatched as three findings, because they have three different fixes.
A workspace can hold client stores with their own access requests, pooled credits and per-client reporting. It is opt-in per workspace and off by default: reaching another merchant's store is not a switch a workspace owner should be one click away from.
The launch engine takes a niche, researches a name with a free matching domain, provisions the accounts the store needs and works through the build as a sequence of steps you can watch and resume. Where a step ends at a person — accepting an invitation, adding a card — it says so instead of reporting itself finished.
A thumbs-down removes that reply from the examples the agent learns your voice from, immediately, rather than ranking it lower and leaving it in the corpus. Writing the reply that should have gone out instead teaches the desk directly; the correction is offered and never required.
The command palette (Cmd+K) now uses fuzzy search across many more page types and entity kinds, so you find what you need with fewer keystrokes. Results are grouped by category, and a curated zero-state shows your most useful pages when nothing is typed. Quick actions for dark mode, feedback, help, changelog, and sign-out are also available directly from the palette.
The Marketing Brain now runs on Claude Sonnet 5, Anthropic's latest model. You can expect sharper campaign proposals and more relevant recommendations when the Brain generates your marketing strategy.
You can now mark each log entry as a Refund (money back, item kept) or a Return (item shipped back), with carrier and tracking fields that appear only when relevant. A new All / Refunds / Returns filter and per-row badges make it easy to spot what came back. You can also log several products in a single record using the new Add Product rows, and the detail page shows a full per-product breakdown with totals.
You can now export four new entity types from the bulk configurator: Files (images, videos, and generic files with URLs and alt text), Menus (Online Store navigation flattened to one row per item), Companies (B2B company records with contact details), and Payouts (Shopify Payments history with date, status, and net amount). These are export-only sheets — import and push are not yet supported for these entities.
You can now export four additional entity types from the bulk export tool: store details, draft orders (with line-item rows and date/tag filtering), discount codes and automatic discounts, and metaobjects with their custom field columns. These are available under a new Store group in the export entity picker.
Bulk import and export are now accessible directly from the Products page and the import flow, so you no longer need to hunt through the sidebar. Every entry point opens the right tab automatically, and the same options are also available under Settings for quick access.
When exporting, you can now pick exactly which groups of columns to include for each entity — for example, export Products with just Basic and Inventory data, or Customers with Addresses but without Metafields. The Basic group is always included so your files stay re-importable. Exports with no selection made behave exactly as before.
You can now bulk-export Blog Posts (and import them back) and export your historical Orders as a spreadsheet. Exports can be downloaded as Excel or CSV, and you can filter Products and Orders by status, tag, vendor, date, and more before exporting. The bulk jobs page also shows clearer labels, relative timestamps, and automatically resumes tracking any jobs still in progress when you return to the page.
// direction
// stay current
// changelog email
// one email per ship · unsubscribe is one click · no tracking pixels
// see it run
Connect your store and the agents start in dry-run — you see exactly what every future release would have done before it touches a thing.
Dry-run by default · Every action logged · One-click rollback
We use cookies to run this site, to see which pages help, and to measure which ads bring people here. You choose per category; everything except the strictly necessary is off until you say otherwise. Cookie policy · Privacy
