Shopify launched Canvas on 1 October 2026. You describe the store you want, Sidekick writes the theme code, and every page sits in one workspace while it works. Shopify's own framing is that a merchant can build a fully customised store in about twenty minutes. The number is doing a lot of work in that sentence, and the list of things Canvas does not support at launch is longer and more consequential than most of the coverage has mentioned.
Canvas is a chat-driven design surface for Shopify themes, launched 1 October 2026, powered by Sidekick. It supports Shopify and custom themes only, is desktop-only, and does not yet support third-party themes, app blocks and extensions, markets, translations, rollouts or theme updates. It compresses the design step of a build, not the build.
What Canvas actually is
Canvas is a workspace rather than a page editor. It lays out every page of your store at once, so you can see how the homepage, collection pages, product pages and cart relate to each other, and you direct changes by telling Sidekick what you want in plain language. Sidekick writes the underlying Liquid and CSS, takes screenshots of what it produced, checks the result against your instruction and refines it. Within a session it holds on to the design decisions you have already made, so you are not restating your brand rules at every turn.
That screenshot-and-iterate loop is the technically interesting part, and it is the reason the output is better than the first generation of AI site builders. The agent is not producing code and hoping. It is looking at the rendered page and correcting itself, which is close to what a junior developer does and considerably faster.
Shopify's Director of Product, Ben Sehl, put the headline figure at roughly twenty minutes for a fully customised store, against a conventional timeline measured in weeks. Shopify has also been careful to call Canvas an early product that will not replace the existing online store editor for now, while signalling that it expects Canvas to become the default editor once remaining gaps close.
The launch limitations, in one place
This is the list that determines whether Canvas is relevant to your store this quarter, and it is not in the press coverage in a usable form.
| Constraint | Detail at launch | Who this rules out |
|---|---|---|
| Theme support | Shopify-developed themes and custom themes only. Third-party Theme Store themes unsupported; Shopify says compatibility is in progress | Any store running a popular paid third-party theme, which is a large share of the installed base |
| App blocks and extensions | Unsupported, promised over time | Stores whose product pages depend on review, subscription, bundling or upsell app blocks |
| Markets | Unsupported | Anyone selling in more than one country with market-specific storefronts |
| Translations | Unsupported | Any multilingual store |
| Rollouts | Unsupported | Teams that stage theme changes across multiple stores or environments |
| Theme updates | Unsupported | Anyone relying on upstream theme maintenance, which becomes a real liability over time |
| Device | Desktop only | Merchants who run the business from a phone |
| Plan | Only on plans that include theme customisation. A merchant who moves to a plan without it can view a Canvas-edited theme but cannot edit or publish it | Entry-tier merchants, and anyone who might downgrade |
| Availability | Rolling out to selected existing stores, not all merchants at once | Possibly you, today |
Read the app blocks and markets rows together and a pattern emerges. Canvas is strongest on exactly the stores that are simplest: one market, one language, a light app stack, a Shopify theme. That is a real and large group of merchants. It is not the group that was commissioning multi-week Shopify builds in the first place.
Unpacking "twenty minutes"
The comparison being drawn is twenty minutes against a two-week developer timeline, and it compares two different things. A store build is not one task. Setting the design step next to everything else makes the claim legible.
| Part of a Shopify build | What it involves | Does Canvas compress it? |
|---|---|---|
| Discovery and information architecture | Deciding what you sell, how it is grouped, what the navigation promises | No. Canvas executes a brief; it does not write one |
| Product data and merchandising | Titles, descriptions, variants, metafields, collection rules, imagery | No |
| Theme design and layout | Page structure, type, colour, section composition | Yes, substantially. This is the step Canvas is for |
| App selection and configuration | Reviews, subscriptions, bundles, loyalty, each with its own settings and page blocks | No, and app blocks are unsupported at launch |
| Migration | Products, customers, order history, and the redirect map that protects your rankings | No |
| Markets, currencies, translations | Regional pricing, local payment methods, translated content | No, both unsupported at launch |
| Checkout and payments | Payment methods, taxes, shipping logic, checkout customisation | No |
| Integrations | ERP, 3PL, email platform, analytics, feeds | No |
| Performance and accessibility QA | Verifying speed, layout stability and keyboard access on real devices | Partly. It generates code you still have to check |
| Launch | Redirects live, tracking verified, a working test order | No |
One row out of ten, and it is the row that is most visible and most often mistaken for the whole job. Twenty minutes to a designed storefront is a credible claim. Twenty minutes to a store that can take money reliably is not, and Shopify has not claimed it.
The honest read is that Canvas removes the step where merchants used to get stuck waiting on a designer for a change they could describe in a sentence. That is genuinely valuable. It is also the step that was already the cheapest to buy.
Canvas alone, Canvas plus a developer, or developer-led
A decision tree, in the order the questions actually matter. Stop at the first one that applies to you.
- Running a third-party Theme Store theme? Canvas is unavailable. Either stay with the theme editor or budget a move to a Shopify or custom theme, which is a project in itself and not one to start in Q4.
- Selling in more than one market or language? Developer-led. Markets and translations are unsupported, and these are the areas where a mistake is expensive and slow to find.
- Do your product pages depend on app blocks? Canvas plus a developer. Design the storefront in Canvas, keep app block work in the theme editor, and accept that you are maintaining two surfaces for now.
- Migrating an existing store with existing rankings? Developer-led for the migration, Canvas afterwards for iteration. The redirect map is where redesigns lose revenue, and no design tool touches it.
- Is the storefront's conversion rate something you are accountable for? Canvas plus review. Generate freely, then check the output against a written standard rather than by eye.
- None of the above, one market, light app stack, Shopify theme? Canvas alone is a reasonable call, and probably the best value route to a decent storefront that exists right now.
If you qualify for Canvas, use it on a duplicated unpublished theme first and give Sidekick your brand rules explicitly: type scale, spacing, button treatment, what the hero must contain. The quality of the output tracks the specificity of the brief almost exactly, which is the same thing that was true of briefing a human designer.
The maintenance question nobody is asking
Two of the launch limitations combine into a risk worth naming before you build anything substantial. Canvas supports custom themes and does not support theme updates.
A maintained Shopify theme is a product with a roadmap. Its author ships fixes when Shopify changes something, when a browser release breaks a layout, when a new checkout capability appears. The moment your theme is heavily generated or customised, you are outside that path, and the work of keeping it current transfers to you. That is the familiar trade-off of a custom build, and merchants who commission one usually understand they have signed up for it. A merchant who arrived there in twenty minutes through a conversation very likely has not.
So the question to settle at the start is who owns the theme in eighteen months. If the answer is nobody, the cheapest path is a maintained Shopify theme adjusted lightly, and you take Canvas's convenience without its long tail. If you are going to generate something substantial, treat it like any other custom codebase and put it on a maintenance footing from day one.
Review generated output against a checklist, not by eye
A design surface that produces plausible layouts quickly produces plausible mistakes quickly. The failures that cost money do not look like failures on a desktop screenshot. Six checks worth running on anything Canvas publishes:
- Add to cart above the fold on a real phone, not on a desktop viewport scaled down. This is the single most common regression in any redesign.
- Layout stability on load. A section that shifts as images arrive damages both Core Web Vitals and mis-taps on mobile.
- Page weight and script count against what the theme shipped with. Generated sections can add weight invisibly, and theme speed is a conversion input, not a vanity metric.
- Heading structure and keyboard focus. Visual hierarchy that reads correctly can still produce an unusable tab order and a broken heading outline.
- Structured data intact. Product schema that a generated template dropped will not show as an error anywhere you normally look.
- A real test order through every payment method you offer, after every significant change. Not a preview.
None of that is an argument against the tool. It is the same review any Shopify developer should be doing on their own work, written down so a merchant can run it.
Where this sits in Shopify's week
Canvas did not arrive alone. In the ten days before it, Shopify made merchants discoverable and purchasable inside Meta's Muse assistant by default, and extended WebMCP support to checkout so browser-based AI agents can update order details and complete a purchase with the buyer's authorisation. Taken together, the direction is unmistakable: an agent helps build the store, and an agent may well be the one buying from it.
For merchants the practical consequence is that the storefront is becoming less of a destination and more of a data source. Clean product data, correct structured data and a checkout that behaves predictably matter more in that world than a distinctive hero section does. If you are choosing where to spend the time Canvas saves you, spend it there rather than on more design iterations, and read our note on what agentic checkout does to your measurement before Q4 traffic arrives.
What to ask before you commit
If you are talking to an agency about a Shopify project this quarter, Canvas changes three questions worth asking. Will the storefront be built on a theme Canvas can edit later, so you are not locked out of your own design tooling? Who owns theme maintenance once the build includes generated code? And what is being delivered besides design, since that is now the part you cannot do in an afternoon. A partner who answers those clearly is worth more than one who either dismisses the tool or pretends it does the whole job. We take the same view of it as we do of AI in web development generally: real leverage on the production step, no change at all to the parts that require judgement.
Frequently asked questions
What is Shopify Canvas?
Canvas is a store design surface Shopify launched on 1 October 2026. It shows every page of a store in one visual workspace and lets a merchant direct changes in plain language through Sidekick, Shopify's AI agent, which writes the underlying theme code. Sidekick takes screenshots of its own output, checks the result and iterates, and it remembers design decisions made earlier in the session. It is desktop-only at launch.
Does Shopify Canvas work with any theme?
No. At launch Canvas supports Shopify-developed themes and custom themes only. Themes bought from the Shopify Theme Store by third-party developers are not supported, although Shopify says compatibility is in progress and that Canvas will eventually become the default editor. If your store runs a popular paid third-party theme, Canvas is not available to you yet regardless of your plan.
Can Shopify Canvas really build a store in 20 minutes?
Shopify's Director of Product, Ben Sehl, said merchants can build fully customised stores in around twenty minutes with Canvas. That claim is about the theme design step, which is one slice of a store build. Product data, app configuration, markets and translations, payments, integrations, migration, redirects and QA are untouched by it, and several of those are unsupported in Canvas at launch. Twenty minutes to a designed storefront is plausible. Twenty minutes to a store that can take orders safely is not.
What does Shopify Canvas not support at launch?
Shopify has listed third-party themes, app blocks and extensions, markets, translations, rollouts and theme updates as unsupported at launch, with support promised over time. It is also desktop-only, limited to plans that include theme customisation, and rolling out to selected existing stores rather than every merchant at once. A merchant who downgrades to a plan without theme customisation can still view a Canvas-edited theme but cannot edit or publish it.
Does Shopify Canvas replace the theme editor?
Not yet. Shopify has said Canvas is an early product that will not replace the existing online store editor for now, while also signalling that it expects Canvas to become the default editor once third-party theme compatibility lands. Treat it as a second editing surface for the next few quarters, and keep the theme editor as the place where app blocks, markets and translations are handled.
Do I still need a Shopify developer if I use Canvas?
For a simple catalogue, one market, one currency and a light app stack, Canvas plus a careful merchant can get a credible storefront live. You still need development help when any of the following apply: a third-party theme, multiple markets or languages, app blocks and checkout extensions, a migration with existing URLs to preserve, custom integrations to an ERP or 3PL, or a conversion and speed budget you are accountable for. Canvas generates code; it does not take responsibility for it.
What is the risk of letting AI write your theme code?
The two practical risks are maintenance and measurement. A heavily customised theme sits outside the update path of a maintained Shopify theme, and theme updates are listed as unsupported in Canvas at launch, so you inherit a codebase someone has to keep working. And generated layouts can quietly undo conversion work: a hero that pushes the add-to-cart below the fold or a section that shifts on load costs revenue without throwing an error. Review generated output against a written checklist, not by eye.
The takeaway
Canvas is the most capable version of conversational store design anyone has shipped, and it removes a real bottleneck for merchants who could describe what they wanted but could not build it. It is also narrower than the headline suggests: one theme family, one market, no app blocks, no theme updates, desktop only, on selected stores. If you fit inside those lines, use it, brief it properly and review what it publishes. If you do not, the sensible move this quarter is to make sure your next storefront is built on a theme Canvas will be able to edit when the gaps close, and to spend the saved design time on product data and checkout instead.
Sources & further reading
- TechCrunch, "Shopify debuts Canvas, a way to build online stores by chatting with AI" (1 October 2026)
- Shopify Changelog, Canvas availability, supported themes and plan requirements
- Unite.AI, "Shopify Rolls Out Canvas, a Sidekick-Powered Store Design Surface"
- Common Thread Collective, "Shopify Just Launched Canvas on October 1: What the AI Site Builder Means"
- TechCrunch, "Shopify opens checkout to browser-based AI agents" (28 September 2026)
- PYMNTS, "Shopify Brings Shop Pay Checkout Solution to Meta's Muse AI Agent"