Will Your Apps Survive a Move to Hydrogen?
App compatibility after a Hydrogen migration is a BUILD: theme app extensions integrate with Online Store 2.0 themes, and a Hydrogen storefront has no Liquid theme for them to render into. Each reviews, upsell or personalization widget becomes a React component against the vendor's API, at $2,000–$8,000 per surface (Deploi estimate, illustrative). Keep the subscriptions whose data you need and cancel the ones that sold only a widget.
Your profile — see how the verdict shifts
- Confidence
- High — Shopify's theme app extensions documentation, re-read on 2026-09-05, states that theme app extensions can integrate with Online Store 2.0 themes. That page says nothing about Hydrogen or headless storefronts, so the consequence is ours rather than Shopify's: a Hydrogen storefront has no Liquid theme, which leaves app blocks and app embeds without a host. Confidence is high on the incompatibility itself and lower on the absence of any bridge, because we did not exhaustively test every extension type. The commercial side was verified the same day on three live listings a mature Plus stack usually contains: Judge.me (5.0★, 46,378 reviews), Loox (4.9★, 9,515 reviews) and Rebuy Personalization Engine (4.7★, 848 reviews), all Built for Shopify. Every one of them delivers its storefront output through the theme, and every one of them also exposes data and APIs, which is what makes this a front-end rebuild rather than a replacement project. The important number is the count of rendered surfaces, not the count of installed apps: a single reviews app commonly renders on the product page, the collection card, the cart, the homepage and a dedicated reviews page.
- Reference scenario
- $20M–$100M GMV · Shopify Plus · mature Online Store 2.0 app stack moving to Hydrogen · agency or in-house React bench
- As of
- September 2026
Decision at a Glance
| Your profile | Verdict | Why |
|---|---|---|
| Under 5 storefront-rendering apps | BUILD | The rebuild is a sprint, not a project. Take the chance to drop the apps whose only contribution was a badge, and keep the two that hold real data. |
| 5–15 storefront-rendering apps | BUILD | This is the typical mature stack, and the rebuild is the largest hidden line in a Hydrogen estimate. Inventory it before signing anything, not during week three. |
| More than 15 rendered surfaces | DEPENDS | At this density the rebuild can outweigh the storefront itself. Phase the migration by template, or reconsider whether the storefront problem really requires going headless. |
| No React bench | WAIT | Online Store 2.0 is where theme app extensions render and where your stack already works. Without a bench to own React components, the migration converts working features into tickets. |
What App Compatibility After Headless Migration Actually Drives
| Outcome | Impact | How it works |
|---|---|---|
| Customer experience | High | Reviews, upsells and recommendations vanish from product pages the day the theme does, which is the most visible regression a headless cutover produces. |
| Revenue — direct | High | Upsell and cross-sell placements convert on their own, so a missing widget is a measurable revenue line rather than a cosmetic gap in the design. |
| Revenue — indirect | Medium | Star ratings and review content feed both conversion and organic snippets, so losing them on a rebuilt product page costs traffic as well as trust. |
| Operational efficiency | Medium | Marketing loses the theme editor it used to place widgets, so every placement change becomes a developer ticket unless those controls get rebuilt. |
| Data & insight | Low | The vendor keeps the underlying review and behavioral data through the migration; what changes is who decides where it appears. |
Spend ceiling: The subscriptions are the cheap part of this decision. Size the budget to the number of storefront surfaces each app renders, because that count is what the rebuild actually costs. A stack of 12 apps drawing 20 surfaces is a bigger line than the homepage, and it is the line most headless proposals never show.
What buying enables (top apps)
- + Review collection, moderation, syndication and rich snippets maintained by a vendor, on free tiers that carry real order volume
- + Personalization and recommendation logic tuned by a team whose entire product is that logic, with placements a merchandiser can move
- + Documented APIs from the larger vendors, which is what keeps this a front-end rebuild rather than a full replacement project
What building additionally unlocks
- + Every storefront surface rendered by your own code in your design system, with no injected vendor script in the critical path
- + Server-rendered reviews and recommendations, which is where a headless storefront's performance advantage actually comes from
- + Consolidation: one component reading one API can replace three widgets from three vendors, and three subscriptions with them
- + Placement logic held in your own content model, so merchandising changes ship without waiting on a vendor's editor
Find Your Verdict in 3 Questions
Does your team have a React bench to own a Hydrogen storefront?
Yes: Go to question 2.
No: Your verdict: WAIT — stay on Online Store 2.0, where theme app extensions render and the stack you pay for keeps working.
Does each storefront-rendering app expose an API your Hydrogen code can call?
Yes: Go to question 3.
No: Your verdict: BUILD — replace those apps, because a widget with no API has nothing left to offer a Liquid-free storefront.
Is the data the app holds worth keeping, such as review history or referral records?
Yes: Your verdict: BUILD — keep the subscription for the data and rebuild the surface as a React component.
No: Your verdict: BUILD — cancel the subscription and build the surface directly, because the widget was the whole product.
The TCC Scorecard — 12 Dimensions
TCC — Total Cost of Capability: what it actually costs to have this capability over three years, whichever way you get it. Each dimension is scored 0–5 for both paths. How we score →
| Dimension | Buy | Build | Why |
|---|---|---|---|
| Cost | |||
| Acquisition & implementation | Keeping the apps costs nothing at migration time, right up to the day the storefront renders without them. | ||
| Recurring fees | Judge.me runs free to $15/month, Loox $49.99–$299.99/month and Rebuy from about $25/month scaling with orders (verified Sep 2026); a rebuilt surface adds no new subscription. | ||
| Maintenance & upgrades | The vendor maintains its own widget; your React component is yours to keep working against their API forever. | ||
| Switching & exit | Reviews, referral records and personalization models sit in the vendor's database, so leaving is an export question either way. | ||
| Risk | |||
| Vendor risk | Review bases here are enormous at 46,378, 9,515 and 848, so the risk is rendering rather than vendor solvency. | ||
| Security & compliance surface | A theme app extension injects vendor script into your storefront; a component you write calls the API from code you control. | ||
| Platform-deprecation exposure | Theme app extensions are tied to Online Store 2.0 themes, which is precisely the dependency a headless migration removes. | ||
| Value | |||
| Fit to requirement | A rebuilt component matches your design system and your layout; the vendor widget matches the vendor's, wherever it lands. | ||
| Time to market | Keeping an app is instant; rebuilding ten surfaces is the long pole of a Hydrogen cutover and the reason launches slip. | ||
| Performance & scale | Removing injected third-party scripts is one of the real performance wins of headless, and rebuilt components render server-side. | ||
| Data ownership & AI-readiness | The vendor still holds reviews and behavioral data; what you gain is control over where, when and how it appears. | ||
| Focus & opportunity cost | Rebuilding widgets produces no new capability and cannot be skipped, which is the worst combination on a roadmap. | ||
The App Landscape
| App | Status | Pricing | Best for |
|---|---|---|---|
| Judge.me Product Reviews App | Live — 5.0★, 46,378 reviews; Built for Shopify. The largest review base on the App Store and the most common reviews app in a mature Plus stack. Its storefront output arrives through theme widgets, so every star rating, carousel and review section is a surface to rebuild. The data, importer and integrations are the reason to keep paying after the migration. | Free (unlimited product and store reviews, widget, star rating badge, carousels, Google reviews sync, trust badge and importer); Awesome $15/month (AI replies, summaries and translations, 130+ integrations, 16 widgets, coupons and referrals, Google Shopping and social sync); 15-day free trial (verified Sep 2026) | Keeping review data and syndication while you rebuild the display yourself |
| Loox - Product Reviews App | Live — 4.9★, 9,515 reviews; Built for Shopify. Photo and video reviews with referrals and syndication. Pricing is metered by orders rather than by widgets, so the bill continues at full rate on a storefront that no longer renders its 17-plus widgets. Confirm the API surface before assuming each of those placements can be rebuilt. | Free (core review features, capped at 500 orders/month); Convert $49.99/month (300 orders included, $50 per additional 300, AI sorting and summaries, video reviews, referrals, API access); Unlimited $299.99/month with priority support; 14-day free trial on Convert and Unlimited (verified Sep 2026) | Photo and video review programs where the content, not the widget, is the asset |
| Rebuy Personalization Engine | Live — 4.7★, 848 reviews; Built for Shopify. Cart, merchandising, post-purchase and search personalization, priced by orders per month. Personalization is the category where the rebuild hurts most, because the widget is also the placement logic: rebuilding the component means rebuilding where and when it appears, not only how it looks. | Free for development stores; Rebuy Monetize free, paying the merchant $0.20–0.35+ per order; Standard starts as low as $25/month and scales with orders per month; 14-day free trial (verified Sep 2026) | Merchants keeping the recommendation engine while owning every placement in React |
| Rebuilt storefront components (build lane) | Build lane — Each surface an app used to render, written as a React component in the Hydrogen storefront against the vendor's API, or replaced outright where no API exists. Server-render reviews and ratings so the SEO value survives the migration, and put placement rules somewhere a merchandiser can reach. Count surfaces, not apps: one reviews app routinely renders five. | $2,000–$8,000 per storefront surface (Deploi estimate, illustrative) | Every app whose product is data plus an API, which is most of the good ones |
The Build Path
- Inventory the stack by rendered surface, not by app: Open the theme's app blocks and app embeds and list every place a vendor draws something: product page, collection card, cart drawer, homepage, account area, footer. That list is the estimate. The installed-app count is not, and using it is how a migration budget ends up wrong by a factor of three.
- Sort the stack into keep, rebuild and cancel: Keep the vendors whose product is data and an API, and keep paying them. Rebuild the interface each one used to render. Cancel the apps whose only product was the widget, because a Liquid-free storefront leaves them nothing to do. The cancel pile usually pays for a meaningful share of the rebuild.
- Rebuild against each vendor's API, server-side where it matters: Server-render reviews and ratings so structured data and organic snippets survive the cutover. Fetch recommendations at the edge with a cached fallback, so a slow vendor response degrades to a sensible default instead of an empty block. Avoid recreating the client-side script waterfall you just removed.
- Give merchandising its controls back: The theme editor is gone, and with it the way marketing moved a widget without a developer. Placement, copy and toggles need a home: metaobjects, a headless CMS or a small settings surface. Skip this and every placement change becomes a deployment, which is how a headless launch turns into a ticket queue.
- Effort band
- $2,000–$8,000 per storefront surface, typically $20,000–$90,000 across a mature stack — Deploi estimate (illustrative); lands in the $25–75K contact-form band for most stacks and above it past roughly 15 surfaces
- Typical timeline
- 4–10 weeks running in parallel with the storefront build (Deploi estimate, illustrative)
- Maintenance, honestly
- ~15–20% of build cost per year (Deploi estimate): roughly $3,000–$18,000/yr (Deploi estimate, illustrative) tracking each vendor's API changes, which become your integration to maintain rather than theirs.
- What you own — and what you take on
- You own: every storefront surface an app used to render, and the API integrations behind them. You take on: each vendor's API versioning, and the merchandising controls the theme editor used to supply for free.
3-Year Total Cost of Capability
| Buy (app path) | Build (custom path) | |
|---|---|---|
| Year 0 (setup) | $0 (apps already installed) | $20,000–$90,000 |
| Years 1–3 (recurring) | $1,800–$21,600 (app subscriptions) | $9,000–$54,000 (maintenance, retained data subscriptions excluded) |
| 3-year total | ≈$1,800–$21,600 | ≈$29,000–$144,000 |
- † All figures illustrative samples for the reference scenario — not quotes, not verified pricing.
- † Left column is the app stack you already pay for, held flat; it is what the subscriptions cost, not an alternative that keeps the widgets rendering.
- † Right column is the rebuild a Hydrogen migration forces, at roughly 10 rendered surfaces, plus upkeep; three-year horizon.
What the Sticker Price Hides
On the buy path
- — Keeping the subscription does not keep the widget: theme app extensions integrate with Online Store 2.0 themes, and Hydrogen has no theme
- — Loox's Convert plan includes 300 orders and adds $50 per additional 300, so the bill tracks order volume rather than the widgets you can no longer render (verified Sep 2026)
- — Vendor headless support varies and most listings do not state it, so confirm the API surface before you assume a rebuild is even possible
- — Tiers bought for widget features stop earning their price the moment your own code renders the surface
On the build path
- — The estimate scales with rendered surfaces rather than installed apps, and one app commonly renders four or five
- — Merchandising loses the theme editor, so placement controls must be rebuilt in metaobjects or a headless CMS or every change becomes a deploy
- — Each vendor's API becomes your integration to version, which is maintenance the subscription used to absorb
- — ~$3,000–$18,000/yr upkeep (Deploi estimate, illustrative)
What Merchants Say
Teams describe the same discovery mid-migration: the storefront renders beautifully, and the product page is missing reviews, badges, recommendations and an upsell block nobody had listed as a dependency.
The recurring frustration is scope found late rather than scope being large: an app inventory taken in week two would have moved the estimate before the contract instead of after it.
If You Change Your Mind Later
If you bought and outgrow it
Keeping every app costs nothing new and delivers nothing on a Hydrogen storefront, which is a trap rather than an exit. Cancel deliberately, after exporting reviews, referral records and personalization configuration, because those exports are what a rebuilt surface has to be seeded with on day one.
If you built and want out
Rebuilt components are your code against documented APIs, so changing a review or personalization vendor becomes a component swap instead of a theme migration. What stays with the vendor is the accumulated data, which is exactly why the export question belongs at signature rather than at cancellation.
When This Answer Changes
We're watching for:
- ▸ Vendors publishing first-class headless SDKs or Hydrogen components, which would turn part of this rebuild back into an install
- ▸ Shopify shipping any rendering bridge between app extensions and headless storefronts, which does not exist today
- ▸ Your own stack passing roughly 15 rendered surfaces, the point where the rebuild line outgrows the storefront line it sits beside
Verdict change log:
No changes since first publication (September 2026).
Common Questions
Do theme app extensions work on a Hydrogen storefront?
No. Shopify documents that theme app extensions integrate with Online Store 2.0 themes, and a Hydrogen storefront has no Liquid theme to integrate with. App blocks and app embeds therefore have nowhere to render. The apps keep running and keep billing, from free to $299.99/month across the three largest review and personalization vendors (verified Sep 2026), while their storefront output disappears.
Which apps actually break, and which are fine?
Apps break when their product is a rendered widget, and survive when their product is data plus an API. Reviews, upsells, badges, personalization blocks and cross-sell carousels all render through theme app extensions and all stop appearing. Email, analytics, ERP connectors, shipping and back-office apps keep working, because they never touched the theme. Inventory the 2 groups before scoping.
What does rebuilding the app surfaces cost?
Rebuilding costs $2,000–$8,000 per storefront surface, which lands between $20,000 and $90,000 across a mature stack, plus roughly 15–20% of build cost per year in upkeep (Deploi estimate, illustrative). Count rendered surfaces rather than installed apps, because one app often renders four or five. Judge.me runs free to $15/month and Loox $49.99–$299.99/month (verified Sep 2026), so the subscriptions are never the expensive half.
Your Next Steps
If you're going with BUILD(matches your selected profile)
- Open the theme's app blocks and app embeds and list every surface a vendor renders, template by template
- Sort each app into keep, rebuild or cancel, and total the cancel column against the rebuild estimate
- Confirm each retained vendor's API and rate limits before promising the component that depends on it
- Server-render reviews and ratings so structured data and organic snippets survive the cutover
- Rebuild merchandising controls in metaobjects or a headless CMS, so placement changes do not need a deploy
If you're going with BUY
- Keep the vendors whose data you would have to migrate, and negotiate the tier down once the widgets are yours
- Export reviews, referral records and personalization configuration before any cancellation date
- Ask each vendor directly whether it publishes a headless SDK, since most listings do not say
- Re-price the stack after cutover, because tiers bought for widget features no longer earn their cost
Official Docs & Sources
- Theme app extensions — shopify.dev
- Migrate to Hydrogen — shopify.dev
- Hydrogen — shopify.dev
Official documentation linked for verification — our verdicts and estimates are our own.
Related Decisions
Shopify Plus Multipass or a Modern SSO App?
Multipass needs legacy customer accounts, deprecated 2026-02-26 with the sunset date pending, and the only Shopify SSO app is delisted. Connect your IdP instead.
Should You Run Hydrogen and Your Theme in Parallel?
Hydrogen and an Online Store 2.0 theme can run against one catalog and cart. Shopify documents the compatibility; the routing layer between them is yours to build.
Plus's 100 Themes vs. a Multi-Tenant CMS for Sub-Brands?
Plus allows up to 100 themes per account, so theme count rarely decides anything. What decides sub-brand storefront architecture is who maintains the divergence.
Should You Go Headless on Shopify or Stay on Your Theme?
A headless storefront pays off for a small minority of mid-market Shopify merchants; a well-built theme delivers most of the gains without the replatform.
Build or Buy Performance Monitoring & App Audits on Shopify?
Measurement is free on Shopify; storefront speed comes from an audit-and-remediation program, not a speed app.
Counting the app surfaces before a Hydrogen cutover?
We inventory every place an app renders today, sort the stack into keep, rebuild and cancel, and put a number on the line most headless estimates leave out.
Contact us todayVerdict scored for the reference scenario above. Estimates are not quotes; app pricing carries its verification date and gets re-verified quarterly. Full scoring anchors: see the TCC methodology.
Read how we score these decisions (the TCC Framework). No affiliate links, no paid placement — no app vendor pays to appear here.