
On this page
- Shopify Liquid vs Hydrogen: the short answer
- What you actually own with each
- Cost: where the money goes on each
- Hydrogen vs Liquid performance: Liquid starts ahead
- Headless Shopify SEO: what you inherit and what you rebuild
- Team skills, apps and the work nobody demos
- When each one fits: a weighted scorecard
- The middle paths most comparisons skip
- Frequently asked questions
- What we’d do first
For most Shopify brands in late 2026, build on Liquid. A well-built Liquid theme is cheaper to launch and run, starts faster in real-user data, gives you sitemaps and SEO plumbing for free, and lets your marketing team ship pages without a developer. Hydrogen earns its cost when one Shopify backend has to feed several front ends, or when the storefront needs to behave like an application and you can pay React engineers year after year.
That is the whole Shopify Liquid vs Hydrogen decision. Below is the evidence, factor by factor: cost, speed, SEO, the team you need, and the daily work nobody demos. Plus the in-between options that opened up when Shopify turned Hydrogen into a library in October 2026.
Key takeaways
- Liquid means Shopify renders, hosts and caches your storefront. Hydrogen means your team owns a JavaScript application and Shopify supplies data and checkout.
- Hydrogen 2026-10, released October 14, 2026, is now a library you add to React Router, Next.js or other frameworks.
- In Shopify’s own real-user analysis, 59.5% of Liquid origins passed all Core Web Vitals against 35% of Hydrogen origins. Headless has to earn its speed through caching work.
- Liquid stores get
sitemap.xml,robots.txtand app blocks without writing code. On Hydrogen, meta tags and structured data are your responsibility. - Choose by the team you can keep, not the demo you saw. Hydrogen without two senior front-end engineers tends to decay.
Shopify Liquid vs Hydrogen: the short answer
Pick Liquid unless you can name a requirement a theme cannot meet. Hydrogen is not a better Liquid. It is a different ownership model that pays off for a specific kind of business.
Here is the split we see across the builds we scope:
| If this describes you | Build on | Why |
|---|---|---|
| Single web store, marketing team publishes weekly, relies on 10+ storefront apps | Liquid | Theme editor and app blocks do the work a headless team would otherwise build |
| Wants a store that looks less like a template, or loads faster on mobile | Liquid (custom theme) | Design and speed problems are usually theme and app problems, not architecture problems |
| Already runs a Next.js or Nuxt content site and wants commerce inside it | Hydrogen (as a library) | You add cart and product logic to a codebase you already maintain |
| Sells through web, a native app and in-store screens from one front-end codebase | Hydrogen | One API-driven front end shared across surfaces is what headless is for |
| Main frustration is checkout | Neither choice fixes it | Both send buyers to the same Shopify checkout |
That last row catches people. Shopify’s Hydrogen documentation is plain about it: Hydrogen ships no checkout UI, because checkout stays on Shopify. Going headless to fix checkout leaves the problem where it was.
What you actually own with each
With Liquid, Shopify owns rendering, hosting, CDN and caching, and your team customizes templates inside that system. With Hydrogen, your team owns routing, rendering, data loading and caching; Shopify keeps data and checkout. Everything else follows from that.
Look at the same job in both. A title, price and add-to-cart button in a Liquid section:
<h1>{{ product.title }}</h1>
<p>{{ product.selected_or_first_available_variant.price | money }}</p>
{% form 'product', product %}
<input type="hidden" name="id" value="{{ product.selected_or_first_available_variant.id }}">
<button type="submit">Add to cart</button>
{% endform %}
The product object is already there. Currency formatting, the cart endpoint and the page’s hosting are Shopify’s problem. In a headless build, step one is asking the Storefront API for that data:
query Product($handle: String!) {
product(handle: $handle) {
title
seo { title description }
selectedOrFirstAvailableVariant {
id
price { amount currencyCode }
}
}
}
Then your framework renders it, Hydrogen’s cart (in React, CartProvider and useCart) handles add-to-cart, your code sets the cache rules, and your route sets the meta tags. None of it is hard for a good React team. All of it is yours to maintain.

What changed with Hydrogen 2026-10
Until this autumn, Hydrogen’s cart meant building the whole storefront on its React Router framework. Shopify’s 2026-10 release notes describe the new version as a library that works with any JavaScript framework. It handles cart, variants, Markets pricing, consent-aware analytics, Shop Pay and customer accounts; your framework does the rest. The starters are a React Router template you deploy to Oxygen from the admin, and Vercel Shop for Next.js. Existing Hydrogen storefronts keep working. The practical effect: a partial move no longer means replatforming everything.
Cost: where the money goes on each
Liquid is cheaper on every line except licence fees, where both are near zero. Hydrogen is free, and Oxygen hosting is included at no extra charge on Starter, Basic, Grow, Advanced, Plus and Pause and build plans, according to Shopify’s Hydrogen and Oxygen fundamentals. (Oxygen is not available on the Agentic plan.) The Shopify Hydrogen cost is engineering time, and it keeps arriving after launch.
| Cost line | Liquid theme | Hydrogen storefront |
|---|---|---|
| Hosting and CDN | Included in your Shopify plan | Oxygen included on paid plans; Vercel or others bill separately |
| Content editing | Theme editor included | Headless CMS subscription, or an editor you build |
| Storefront apps | Most install as app blocks or embeds | Each one needs API work, a replacement or removal |
| Build effort (our planning bands) | 120 to 250 h customized premium theme; 300 to 600 h custom theme | Roughly double the front-end effort of a comparable custom theme |
| Ongoing | Theme updates, app changes | Framework, Hydrogen and API version upgrades, plus monitoring |
A hypothetical example. Say a brand scopes a custom Liquid theme at 400 hours and the equivalent Hydrogen build at 1,000 hours, both at $120 an hour. That is $48,000 against $120,000 to launch. Add ongoing work of 12 hours a month on Liquid and 35 on Hydrogen, plus an assumed $300 a month for a headless CMS, and year one comes to about $65,300 for Liquid and $174,000 for Hydrogen. The gap does not close in year two, because it is mostly those 23 extra hours a month.
We built a fuller three-year version of this model, with the revenue lift headless needs just to break even, in our guide to headless Shopify with Hydrogen. Below a few million dollars in online revenue, the required lift gets hard to believe.
Watch out: a Hydrogen quote that lists build hours but no monthly retainer is not a cheaper quote. It is a quote for a storefront that will be out of date within a year. Ask every vendor what the second year costs before you compare totals.
Hydrogen vs Liquid performance: Liquid starts ahead
Liquid stores are faster by default, and Hydrogen matches or beats them only with deliberate caching work. That is roughly what Shopify’s own performance team found in real-user data.
In Shopify’s analysis of Chrome UX Report data, 59.5% of origins built on Shopify Liquid passed all Core Web Vitals in September 2023, against 35% for Hydrogen and 43% for all websites. Next.js and Remix sites sat around 28% to 29%, and Nuxt at 22%. On responsiveness the two Shopify options were neck and neck: 75% of Liquid origins and 74% of Hydrogen origins had good INP.

Caveats first: the data predates Hydrogen 2026-10, the Hydrogen bucket mixes stores with non-commerce sites, and correlation is not causation. The author’s conclusion is the useful part: an unoptimized headless site is likely to be slower, and Hydrogen only pulls ahead with full-page and sub-request caching, which takes a dedicated developer team. We found no newer public dataset splitting the two, so ask any vendor quoting one for the method.
Why does Liquid start ahead? Shopify hosts, renders and caches the theme for you, and a theme like Horizon ships with years of performance tuning. On Hydrogen, your code decides what gets cached and for how long. Oxygen adds hard limits to design around: a 10 MB worker, 400 ms startup, 30 seconds of CPU per request, 128 MB of memory and no proxies in front.
On most slow stores we audit, the culprits are app scripts, oversized hero images and third-party tags, and those follow you to headless. Fix them first with our Shopify speed optimization guide, then decide whether you still have a speed case for Hydrogen.
Headless Shopify SEO: what you inherit and what you rebuild
Headless Shopify SEO is not worse by nature, but nothing comes free. A Liquid store inherits Shopify’s SEO plumbing on day one. On Hydrogen, every item your team forgets becomes a ranking problem you find months later.
On Liquid, Shopify generates sitemap.xml automatically, with child sitemaps for products, collections, blogs and pages, updated whenever you add content. International domains each get their own sitemap and extra languages are added for you. robots.txt is generated too, and if you need custom rules you add a robots.txt.liquid template that loops over robots.default_groups so you keep Shopify’s defaults while adding your own.
On Hydrogen 2026-10, the docs say meta tags, Open Graph tags and structured data are your responsibility, via your framework’s head or metadata API. The checklist we run on every headless build:
| SEO item | Liquid theme | Hydrogen storefront |
|---|---|---|
| XML sitemaps | Automatic, including markets and languages | Build sitemap routes and keep unpublished items out |
| robots.txt | Automatic; override with robots.txt.liquid |
Build a route; block preview deployments |
| Title, description, canonical | Theme outputs them from admin SEO fields | Each route must query and render them |
| Product structured data | Most modern themes output it | Write and maintain JSON-LD yourself |
| International domains and languages | Sitemap per domain, languages added automatically | Your locale routing must output hreflang and per-market sitemaps |
| URL redirects from admin | Work automatically | Your app must read and honour them |
| Real 404 status codes | Automatic | Easy to get wrong in client-side routing |
| URL structure | /products/handle by default |
Shopify recommends supporting /products/:handle or a server-side 3XX redirect to it |
The bug we find most on headless audits is the soft 404: a deleted product showing “not found” with a 200 status, so it stays indexed. Second place: staging deployments nobody blocked from crawlers.
From our team: if you go headless, crawl the staging site with Screaming Frog before launch and diff it against a crawl of the current Liquid store. Titles, canonicals, status codes and structured data should match or improve, URL for URL. It catches most launch-week ranking drops. Our Shopify SEO guide covers what “good” looks like on each template.
Team skills, apps and the work nobody demos
Liquid needs a Shopify theme developer. Hydrogen needs a small product engineering team. That shapes hiring and marketing speed more than any benchmark.
What each team looks like
A strong Liquid theme developer knows Liquid, HTML, CSS and modern JavaScript, understands section and block schemas, works with shopify theme dev in the Shopify CLI, and knows the theme editor well enough to design sections marketers can use safely. One good developer, plus design support, can run a busy store. Most retainers on our Shopify development team look like that: one lead theme developer, with design and QA as needed.
A Hydrogen storefront needs:
- Senior TypeScript and React (or Vue) engineers who understand server rendering and caching.
- GraphQL fluency against the Storefront API and Customer Account API. Hydrogen does not cover the Admin API, so anything that writes to the store needs separate backend work.
- Someone who owns CI/CD, previews, monitoring and hosting limits.
- A CMS integration good enough that marketers stop filing tickets for banners.
- Analytics and consent, now yours end to end.
Our rule: if you cannot keep at least two senior front-end engineers on the storefront, in-house or on a long retainer, do not go headless. If you are hiring for either path, our guide on how to hire a Shopify developer has the screening questions we use.
Apps and the theme editor
This is where Hydrogen budgets break. Shopify’s theme app extension guidelines say app blocks work only in Online Store 2.0 themes with JSON templates, while app embeds work in all Shopify themes. Both are theme features. On headless, a reviews widget or loyalty panel has nothing to attach to, so each app needs an SDK, an API integration, a replacement or the bin.
A normal week of changes on each, from our projects:
| Task | Liquid theme | Hydrogen storefront |
|---|---|---|
| Swap the homepage hero for a sale | Marketer, theme editor, minutes | Marketer in the CMS, if that block was modelled; otherwise a ticket |
| New landing page from existing sections | Marketer, new template | Marketer, if the CMS page type exists |
| Add a reviews app | Install, drop in the app block | Read the app’s API docs, build a component, test, deploy |
| Launch a new market | Settings > Markets; theme picks it up | Same admin setup, plus locale routing, hreflang and QA in your app |
| Platform upgrade | Theme update, merge custom code | Framework, Hydrogen and API version upgrades |
Horizon is part of why Liquid keeps winning this table. It is Shopify’s free theme, built on nested theme blocks, and the Theme Store listing shows version 4.5.0 shipped on November 10, 2026. That closes much of the design gap that used to push brands toward headless. Liquid is not being retired.
Not sure which side of this line your store sits on?
Send us your store URL, your app list and the one thing your current storefront cannot do, and we will tell you whether a Liquid rebuild or Hydrogen fits, with a rough hour range for each.
Get an architecture recommendation
When each one fits: a weighted scorecard
The answer depends on how much each factor matters to you, so score it instead of arguing. Here is the scorecard we use in discovery, filled in for two hypothetical brands. Scores (1 to 5) are our judgement of each brand’s situation; weights are hypothetical too.
- Brand A: DTC apparel, a three-person marketing team launching weekly, 14 storefront apps, no in-house developers.
- Brand B: consumer electronics, a native app, an editorial site on Next.js, six React engineers on staff.
| Factor | Weight A | Weight B | Liquid (A / B) | Hydrogen (A / B) |
|---|---|---|---|---|
| Build and running cost | 25 | 10 | 5 / 5 | 2 / 2 |
| Speed you get without specialist work | 15 | 10 | 4 / 4 | 2 / 4 |
| SEO effort and risk | 20 | 15 | 5 / 5 | 3 / 3 |
| Fit with the team you have | 15 | 15 | 5 / 3 | 2 / 5 |
| Marketing self-serve | 20 | 10 | 5 / 4 | 3 / 4 |
| Front-end freedom and multiple surfaces | 5 | 40 | 3 / 2 | 5 / 5 |
| Weighted score out of 100 | 95 / 66 | 51 / 84 |
Multiply each weight by its score, add them up and divide by five. Brand A is not close: Liquid wins by 44 points. Brand B flips because one factor, sharing a front end across its app and editorial site, carries 40% of the weight, and because its engineers already know React. That is also why B’s Hydrogen speed score is 4, not 2: it has people to do the caching work.

Run it with your own weights before you talk to vendors. If Hydrogen only wins when “freedom” outweighs what your roadmap needs, you have your answer.
The middle paths most comparisons skip
You do not have to choose all of one or all of the other. Since 2026-10, two in-between setups are realistic, and for many brands one beats both extremes.
- Liquid store, Hydrogen inside your existing site. If your editorial site already runs on Next.js or Nuxt, add Hydrogen’s cart and product logic for shoppable content and keep the main store on Liquid.
- Liquid now, headless prototype on the side. Oxygen works on development stores (behind a store login). Build the hardest page, usually the product page, and measure it before anyone signs a rebuild contract.
AI shopping agents no longer decide anything. Shopify says Hydrogen uses the same standard storefront actions as Liquid themes, so in WebMCP-enabled browsers agents can search and update the cart on either.
If you do go fully headless, run it as a staged build: name the one requirement Liquid cannot meet, prototype it, map every app, design the CMS content model before templates, and launch behind an A/B split. That is how our Hydrogen development services team runs these projects, and it is also how we talk clients out of headless when the prototype does not justify it.
Frequently asked questions
Is Hydrogen better than Liquid?
No, Hydrogen is not better than Liquid in general. It is better for a narrow set of brands: those feeding several front ends from one Shopify backend, or building app-like storefronts, with senior React engineers to maintain them. For a single web store, Liquid is cheaper, faster by default in Shopify’s real-user data, and easier for marketers to run day to day.
Does headless Shopify help SEO?
Not by itself. Headless Shopify can rank as well as a Liquid store, but it does not rank better just for being headless. Liquid gives you automatic sitemaps, robots.txt and SEO fields in templates. On Hydrogen, your team must build sitemap routes, meta tags, structured data, hreflang and correct 404s. Done well, results match. Done carelessly, rankings drop after launch.
Is Shopify Hydrogen free to use?
Yes, Hydrogen is free and open source, and Oxygen hosting is included on paid Shopify plans from Starter through Plus, though not on the Agentic plan. The real cost is engineering. In our scoping, a Hydrogen storefront needs about twice the front-end build effort of a comparable custom theme, plus a headless CMS and a monthly development retainer to keep it current.
Can I move from Liquid to Hydrogen later?
Yes, and starting on Liquid keeps that door open. Products, customers, orders, metafields and checkout live in Shopify, not in the theme, so a later Hydrogen storefront reads the same data. What you rebuild is the front end: templates become components, theme settings move to a CMS, and every storefront app needs an API-based replacement.
Is Shopify replacing Liquid with Hydrogen?
No, Liquid is not being replaced. Shopify keeps shipping theme features, and its free Horizon theme received version 4.5.0 on November 10, 2026. Hydrogen and Liquid also share the same standard storefront actions for apps and AI shopping agents. Shopify treats them as two supported ways to build a storefront, not an old and a new one.
What we’d do first
Write one sentence describing what your current storefront cannot do. Not “feel more premium”. A specific capability. Then score it.
If it is about design, speed or conversion, put the budget into a better Liquid theme and an app audit. If it is checkout, look at checkout extensibility. If it is one front end across an app, a content site and the store, and you have the engineers, prototype Hydrogen and decide with real numbers.
We have built Shopify storefronts for brands like Delsey, EcoFlow, Kerbholz and Maker & Moss, on both sides of this line. Our shopify development services cover Liquid themes and Hydrogen builds, so we gain nothing by pushing the expensive one. Bring us that sentence.




