Meet the oBundle Team

Let’s get connected!

Book a time
Stencil vs Catalyst: How to Choose Your BigCommerce Storefront in 2026

Stencil vs Catalyst: How to Choose Your BigCommerce Storefront in 2026

If you run a BigCommerce store, you have probably had a version of this conversation in the last twelve months. Someone, maybe an agency, maybe a developer on your team, maybe a competitor's slick new site, has raised the question of Catalyst. And the pitch usually arrives with a quiet implication attached: that Stencil is yesterday's technology and you are already behind.

That framing is wrong, and acting on it is expensive. The honest version of Stencil vs Catalyst is far less dramatic and far more useful: these are two supported front ends sitting on the same commerce backend, built for different teams with different constraints. One of them is right for you. Quite often it is the one you are already on.

This piece is the decision framework we actually use with merchants before we quote a BigCommerce Catalyst migration. It covers what Catalyst gives you, what you give up to get it, the one structural constraint that blindsides most projects, and four questions that will tell you in about twenty minutes whether you should be having this conversation at all.

First, the part nobody selling a migration leads with

Stencil is not being sunset. There is no announced end-of-life for Stencil themes, and as of this writing BigCommerce's own deprecations and sunsets documentation lists nothing of the kind. The only Stencil entry on that page is narrow and old: support ended for Stencil CLI versions 6 and below on 3 June 2024. That is a tooling version cutoff, the sort every platform issues routinely. It is not a countdown clock on your storefront.

Meanwhile, BigCommerce has continued shipping improvements that land on Stencil stores. The storefront crawler detection list, which drives the is_crawler value themes rely on, was refreshed in September 2026 to recognize newer bots and to correctly classify in-app browsers like WeChat as shoppers rather than crawlers. That is active maintenance of the Stencil surface, not abandonment.

So the deadline pressure is manufactured. You are not choosing between a dying platform and a living one. You are choosing between two live options, which means you get to choose on merit.

What Catalyst actually is

Catalyst is an open-source storefront framework that BigCommerce builds and maintains. It is a Next.js application using React Server Components and Tailwind, which reads commerce data over the GraphQL Storefront API, and which ships with Makeswift, a visual page editor, built in. BigCommerce positions it for teams who want a composable, API-driven architecture and complete control over the front end.

It is under active development. Catalyst 1.12.0 landed in September 2026, alongside CLI 1.4.1, with changes ranging from proxy handling to a meaningful data-cache fix: a correlation header was being attached to cacheable GraphQL requests, generating a unique cache key every time, which meant cached fetches almost never hit. On a product page issuing roughly a dozen queries per view, that is not a rounding error. The fact that this class of bug was found and fixed tells you two useful things at once: the framework is maturing quickly, and it is still young enough that you are closer to the edge than you would be on Stencil.

The thing to hold onto is what does not change. Catalyst replaces the front end. It does not replace BigCommerce.

Stencil and Catalyst architecture compared Stencil renders Handlebars templates on BigCommerce-hosted infrastructure and extends via Page Builder widgets, Script Manager and app scripts. Catalyst runs a Next.js app on hosting you own, reads data over the GraphQL Storefront API, and extends via Makeswift and React components. Both sit on the identical BigCommerce backend for catalog, cart, checkout and orders. Same backend, two different front ends STENCIL CATALYST Shopper's browser Shopper's browser BigCommerce hosting Included. No infrastructure to run. Hosting you own Vercel, Commerce Hosting or similar. Handlebars theme Stencil objects, SCSS, theme marketplace Next.js app, React Server Components Tailwind, your own components, no theme store Page Builder widgets + app scripts Most marketplace apps drop in as-is Makeswift + React components Widgets and Page Builder do not exist here via GraphQL Storefront API The identical BigCommerce backend Catalog · pricing · promotions · cart · checkout · orders · B2B Edition · Multi-Storefront Migrating changes the front end only. Your catalog, orders, customers and checkout stay exactly where they are. Source: BigCommerce developer documentation, "Choosing the right storefront."
Stencil and Catalyst differ above the API line. Everything below it is identical.

Your catalog, pricing, promotions, customer groups, cart, checkout, orders, B2B Edition and Multi-Storefront all stay exactly where they are. This is why a Catalyst project is not a replatform in the way that moving from another platform to BigCommerce is a replatform. It is a front-end rebuild. That is a smaller risk in some ways and a sharper one in others, because everything you are rebuilding is something your customers already use every day.

The five differences that actually decide it

1. Who runs the hosting

On Stencil, BigCommerce hosts your storefront. There is no infrastructure for you to run, patch, scale or pay for separately, and no deploy pipeline to own. On Catalyst, you are responsible for hosting your deployed code. BigCommerce says it is building and optimizing its process around Vercel at present, and there is now a native hosting path with documented GitHub Actions deployment, including pull request previews where the newest open PR claims a shared preview URL.

Framed as a feature, this is control. Framed as an operating reality, it is a standing responsibility: a hosting bill, a deploy process, an uptime owner, and someone whose job it is to care when a build fails at 4pm on a Friday. That is entirely manageable, and plenty of merchants manage it. But it is a new line item in both your budget and your org chart, and it should be named out loud before anyone signs anything.

2. Your app and widget stack

This is the one that kills more Catalyst projects than any other, and it is the one most likely to be glossed over in a sales conversation.

BigCommerce is direct about it. Catalyst is built on the GraphQL Storefront API, which does not support every platform feature, so some things are not available out of the box. Not every App Marketplace app is supported. More pointedly, the guidance for app developers states that widgets and Page Builder functionality in Stencil stores do not exist in Catalyst, and that storefront apps depending on Stencil-specific DOM structure or Stencil Theme Objects are incompatible or only partially functional on headless storefronts unless they ship their own APIs or SDKs. An app whose script does not lean on the Stencil DOM may well be fine, but you confirm that by testing it on a Catalyst storefront, not by assuming.

Translate that into your store. Go and count the things bolted onto your storefront right now: reviews, loyalty, subscriptions, search, upsell widgets, chat, consent management, personalization, that one bespoke script someone added in 2021 that nobody has dared touch since. Each one is a question, and some of the answers will be "this needs rebuilding" or "there is no equivalent." That list, if you compile it honestly, is the real scope of your migration. Everything else is comparatively predictable.

3. Page Builder versus Makeswift

If your marketing team builds and edits landing pages themselves, this difference will be felt daily. Stencil gives them Page Builder and the widget ecosystem around it. Catalyst gives them Makeswift, a drag-and-drop visual editor with multiplayer editing, version history, scheduled publishing and translation tooling.

Makeswift is a capable product, and for some teams it is a straightforward upgrade. But it is a different tool with a different model, and whatever your team has built in Page Builder does not come across. Any custom widget you commissioned is a rebuild as a React component. Budget for retraining, and budget for the quiet period afterwards where the marketing team is slower than they were. Both are real costs and both are temporary, but only if you plan for them.

4. Performance

BigCommerce markets Catalyst with Lighthouse scores of 100 out of the box, and the architecture does support excellent performance. Treat the number carefully, though. A Lighthouse 100 on a reference storefront is a measurement of the starting line. What your shoppers experience is that storefront after you have added your fonts, your imagery, your analytics stack, your consent banner, your chat widget and your tag manager. That is true of any framework. Catalyst gives you a better starting point and better tools; it does not give you immunity from what gets added later.

It is also worth being blunt about the comparison. A well-built Stencil theme that has been properly optimized, with disciplined image handling, a trimmed script payload and a tidy tag manager, performs well. If your current site is slow, there is a strong chance the cause is the thirty scripts on it rather than Handlebars. Migrating to Catalyst while carrying all thirty across solves nothing, and you will have spent six figures to learn that.

5. The team you have, not the team you want

Stencil development means Handlebars, SCSS and JavaScript, plus a theme marketplace you can start from. Catalyst means React, Next.js, TypeScript, GraphQL and Tailwind, with no theme marketplace at all; you customize the provided storefront in code or through Makeswift.

The second skill set is more common in the general job market and scarcer among people who also understand ecommerce specifically. If you have React capability in-house, Catalyst will feel natural and hiring will get easier over time. If your front end is maintained by one contractor who knows Stencil well, moving to Catalyst does not just change your codebase, it changes who can work on your store. That is a strategic decision, not a technical one, and it belongs with whoever owns the budget.

Four gates

Run your situation through these in order. A clear "no" at any gate means the answer is to stay on Stencil for now, and there is no shame in stopping at gate one.

A four-gate decision flow for moving from Stencil to Catalyst Four sequential questions. First, whether you depend on Page Builder widgets or apps tied to the Stencil DOM. Second, whether you have React and Next.js capability plus an owner for hosting. Third, whether your real bottleneck is front-end flexibility Stencil cannot solve. Fourth, whether you can absorb a full rebuild, since no page-by-page migration path exists. A no at any gate means staying on Stencil for now; passing all four means Catalyst is a fit. Four gates. A "no" at any one of them means stay put for now. You're on Stencil, eyeing Catalyst 1. Widgets and apps Do you rely on Page Builder widgets or apps that hook into the Stencil DOM? YES Stay on Stencil Rebuilding them is the whole project NO 2. Capability and ownership React and Next.js skills on hand, plus someone who owns hosting, deploys and uptime? NO Stay on Stencil Hosting you can't staff is a liability YES 3. The actual bottleneck Is front-end flexibility or speed genuinely what's capping revenue, and Stencil can't fix it? NO Optimize Stencil first Cheaper, faster, usually enough YES 4. Appetite for a full rebuild There is no page-by-page path. Can you fund and QA a complete front end before cutover? NO Revisit next budget cycle A half-done rebuild is the worst outcome YES Catalyst is a real fit Scope it as a parallel-channel rebuild Most merchants stop at gate 1 or gate 3. That is a finding, not a failure.
Four gates for a Stencil to Catalyst decision. A no at any gate means stay put for now.

In our experience most merchants stop at gate one or gate three. Gate one stops them because the app stack turns out to be the project. Gate three stops them because, once someone actually measures, the problem they wanted Catalyst to solve was a performance or merchandising problem that Stencil could have solved for a fraction of the cost.

The constraint that catches everyone: there is no gradual path

This deserves its own section because it is routinely assumed away, and because assuming it away wrecks project plans.

You cannot run Stencil and Catalyst side by side on one storefront, serving category pages from one and product pages from the other. A Catalyst maintainer confirmed this directly in the project's public discussions back in July 2024, in response to exactly that question, noting the team was considering migration solutions. The follow-up requests on that thread through 2025, asking for an update or a best-practice guide, have no public answer. As things stand, the only honest position is that incremental, page-by-page migration is not supported.

What this means practically: Catalyst requires its own storefront channel under Multi-Storefront. You build the new front end in parallel, in full, while your existing storefront keeps serving customers. Then you switch.

The parallel-channel sequence for a Stencil to Catalyst migration Six phases: audit apps and widgets; stand up a separate Catalyst storefront channel; rebuild the front end; run parity QA across checkout, promotions, B2B and analytics; prepare SEO with a URL map and redirects; then cut the domain over while the Stencil channel stays available as a rollback. A callout notes there is no supported hybrid in which some pages render from Stencil and others from Catalyst on the same storefront. You don't migrate Catalyst. You build it alongside and switch. No hybrid is supported. You cannot serve category pages from Stencil and product pages from Catalyst on one storefront. It is a full front end, then a switch. 1 Audit Weeks 1–2 Inventory every app, widget, Script Manager entry and integration. Sort into: works as-is, needs rebuilding, has no Catalyst equivalent. 2 New channel Week 2 Catalyst requires its own storefront channel under Multi-Storefront. Your live Stencil storefront keeps running, untouched, the whole time. 3 Rebuild The long pole Components, templates, and Makeswift pages for the marketing team. Budget real time for whatever landed in the "needs rebuilding" column. 4 Parity QA Non-negotiable Checkout, promotions, customer groups, B2B pricing, search, tax, analytics and consent. Test with real accounts, not just the happy path. 5 SEO prep Before cutover Full URL map old to new, 301s staged, structured data and canonicals verified. This is where most replatforms quietly lose their traffic. 6 Cutover One evening Point the domain at the Catalyst channel. Keep the Stencil channel configured and ready for a fortnight. It is the only real rollback you have. Timings are indicative. The audit in phase 1 is what actually sets the length of phase 3.
The parallel-channel sequence. There is no supported hybrid, so the build completes before the switch.

That is a better risk profile than it sounds, because your live store is never in a half-finished state. But it has a hard consequence for budgeting: there is no way to buy a third of a Catalyst storefront, ship it, prove value and fund the rest from the results. You commit to a complete front end before a single customer sees any of it. If your organization cannot fund and QA that in one arc, the right answer is to wait until it can. A stalled rebuild is worse than no rebuild, because you pay for it twice.

If you are staying on Stencil, do these four things

Staying is a legitimate decision, but it should be an active one rather than inertia. Do four things in the next quarter either way, and each one also makes a future migration cheaper:

  • Audit your scripts. Open the tag manager and Script Manager and list everything firing on your storefront. Remove what is dead. This is the single highest-return performance work available to most stores, and it costs almost nothing.
  • Keep Stencil CLI current. Support ended for versions 6 and older in June 2024. Running something ancient is the one place where "Stencil is fine" stops being true.
  • Fix the actual conversion problems. Site search quality, filtering, product data completeness and mobile checkout friction move revenue more reliably than a framework change, and none of them require one.
  • Write the app inventory now. Every app, widget, script and integration, with an owner and a note on whether anyone still uses it. You will need this document the day you seriously evaluate Catalyst, and compiling it calmly beats compiling it under project pressure.

If you are moving, a pre-flight checklist

If you cleared all four gates, these are the things to settle before the build starts rather than halfway through:

  • The app disposition list. Every storefront app sorted into works as-is, needs rebuilding, or has no equivalent, with each one tested on an actual Catalyst storefront rather than assumed.
  • A named hosting owner. A person, a platform decision, a monthly cost, and an escalation path.
  • The URL map. Old to new, complete, with redirects staged and tested before cutover. More replatform traffic is lost here than anywhere else.
  • A parity test plan with real accounts. B2B pricing, customer groups, promotions, tax, gift certificates, search and analytics. Not just the guest happy path.
  • A rollback window. Keep the Stencil channel configured for a couple of weeks after cutover. It is the only genuine rollback available to you.
  • Marketing team enablement. Makeswift training booked before launch, not after the first campaign stalls.

Frequently asked questions

Is Stencil deprecated?

No. BigCommerce's deprecations and sunsets documentation announces no end-of-life for Stencil or Stencil themes. The only related entry is the end of support for Stencil CLI version 6 and older, dated 3 June 2024.

Can I migrate from Stencil to Catalyst one page at a time?

Not in any supported way. A Catalyst maintainer confirmed in the project's public discussions that a hybrid setup, with some page types on Stencil and others on Catalyst, is not currently possible. Plan for a parallel build and a cutover.

Will my BigCommerce apps work on Catalyst?

Some will, some will not. Back-office apps that do not touch the storefront are likely fine. Storefront apps depending on the Stencil DOM or Stencil Theme Objects are incompatible or partially functional, and widgets built for Page Builder do not carry over at all, since Page Builder and widget functionality do not exist in Catalyst. Each app needs testing on a Catalyst storefront individually.

Does Catalyst work with B2B Edition and Multi-Storefront?

Yes. BigCommerce lists B2B, checkout, Feedonomics and Multi-Storefront capabilities as available to Catalyst storefronts, and Catalyst in fact requires its own storefront channel under Multi-Storefront. B2B pricing and permission logic still needs thorough testing against real buyer accounts as part of parity QA.

Will Catalyst make my site faster?

It gives you a faster foundation. BigCommerce cites out-of-the-box Lighthouse scores of 100. What your shoppers actually experience depends on what you add on top. If your current store is slow because of script bloat, that bloat will slow down Catalyst too.

Do I have to use Makeswift?

No, but most teams should. Makeswift is Catalyst's answer to the page-building capability your marketers lose when Page Builder goes away. Skipping it means every landing page change becomes a developer ticket.

Where we land on this

Catalyst is a serious piece of engineering and, for the right merchant, a genuine step change: full control of the front end, a modern React stack, a credible visual editor for marketing, and performance headroom that a heavily customized Stencil theme will struggle to match. For a team with React capability, an appetite to own hosting and a front end that has outgrown Handlebars, it is the right call and we are glad to build it.

For a lot of merchants, though, the honest advice is to stay on Stencil this year, delete half the scripts on the site, fix search and filtering, and revisit Catalyst when the app ecosystem has matured further and the migration tooling the community keeps asking for actually exists. That recommendation does not sell a rebuild. It does tend to make more money for the merchant, which is the part we care about.

If you want a second opinion grounded in your actual app stack rather than a generic comparison, that is the conversation we are always happy to have.