We first published this article in October 2023, and, honestly, it aged badly. Not because it was wrong at the time, but because Shopify has spent the last three years changing almost everything the original relied on. Hydrogen has been rebuilt. checkout.liquid, the legacy system for customising Shopify's checkout pages with code, was fully sunset by August 28, 2025, completing a phased deprecation that had been running since 2023. Its replacement, Checkout Extensibility, uses sandboxed apps and branding APIs instead.

So rather than patch the old post, we've rewritten it from scratch. One thing we've kept, though, is the part readers told us they valued most: a frank section on why you might not want to go headless. We build headless storefronts for a living, so it would be commercially convenient to tell you everyone needs one. They don't, and Shopify's own 2026 guidance now reflects this, noting that headless trades engineering dependency for experience control.

First, a quick recap: what "headless" actually means

Every online store has two halves. There's the back end, the engine room where products, inventory, orders and payments live. And there's the front end, or the "head": everything your customer actually sees and clicks on.

In a standard Shopify Plus setup, both halves come from Shopify. You pick a theme, customise it, and Shopify renders your storefront for you.

Going headless means chopping that head off and building your own. Shopify keeps doing what it's genuinely brilliant at (commerce, checkout, payments, fraud, scale) while your front end becomes a separate application, built with modern web frameworks and talking to Shopify through its Storefront API. An API, if the term is new to you, is simply a structured way for two pieces of software to request data from each other.

Most headless builds also add a headless CMS (a content management system without its own front end) such as Sanity, so marketing teams can edit rich content without touching code. That pairing, Shopify for commerce and Sanity for content, is the combination we reach for most often at Lemon Hive.

What's changed since 2023

A lot. Here's what actually matters if you're weighing this decision today.

Hydrogen grew up

Hydrogen is Shopify's own framework for building headless storefronts. In 2023 it was promising but young. In 2026 it's a genuinely mature option: it now runs on React Router 7, deploys for free to Oxygen (Shopify's own hosting for Hydrogen storefronts), and has moved away from monolithic starter templates towards composable "recipes", smaller ready-made building blocks like subscriptions or B2B features that you add only when you need them. In practice, that means less boilerplate to maintain and fewer painful framework upgrades. We've felt the difference on client projects.

Hydrogen is going framework-agnostic

At Editions Spring 2026, Shopify previewed the bigger shift: Hydrogen's commerce tooling untethered from any single framework. The direction of travel is that you'll be able to use Hydrogen's building blocks with Next.js, Astro, SvelteKit or Nuxt, and deploy to Vercel, Cloudflare or Netlify rather than only Oxygen. As of July 2026, this is still in developer preview, so we wouldn't bet on a production launch just yet, but it signals something important: Shopify is embracing the reality that serious brands want to choose their own stack. Since we build on Next.js and Astro daily (including on Centra and BigCommerce, not just Shopify), we're watching this one closely.

Checkout changed for everyone, headless or not

Checkout. Liquid, the old way of customising Shopify's checkout with code, was fully sunset in 2025. Its replacement, Checkout Extensibility, uses sandboxed apps and branding APIs instead. Similarly, Shopify Scripts (small programs Plus merchants used for custom discounts and shipping logic) reached end of life on 30 June 2026, replaced by Shopify Functions. If you're on Plus and haven't migrated your Scripts, that's now urgent regardless of any headless ambitions.

Why does this matter here? Because checkout customisation used to be a headless talking point, and it no longer really is. Whether your storefront is headless or themed, checkout runs on Shopify's rails and you customise it the same way. In our experience, that's mostly good news: checkout is where money changes hands, and it's the last place you want bespoke code rotting quietly.

Themes got seriously good

This is the change agencies like ours talk about least, so let's be the exception. Shopify's newer theme architecture, with Horizon as the flagship, has closed a large chunk of the gap that used to justify headless builds. Modern themes are fast, flexible and editable by non-developers. Shopify's own 2026 guidance is refreshingly blunt about it: headless trades experience control for engineering dependency, and many stores simply don't need to make that trade.

When the platform vendor itself is telling merchants "you probably don't need this", and we're an agency saying the same, it's worth listening.

The wider platform kept moving

Shopify's momentum hasn't slowed. Shopify accounts for roughly 30 per cent of the US ecommerce platform market (i.e., dedicated ecommerce platforms, distinct from marketplace providers and direct-to-consumer channels) t, and third-party trackers now count tens of thousands of Plus stores, roughly three times the figure we quoted in 2023. Editions Winter 2026 leaned hard into AI: agentic storefronts that surface your products inside ChatGPT, Copilot and Perplexity, theme A/B testing, and per-market checkout customisation. None of that requires headless, and some of it (agentic commerce especially) arrives on themed stores first. That's worth factoring in.

What headless genuinely gets you in 2026

With the hype trimmed away, here's where we still see headless earning its keep, project after project.

Content-led selling

If your brand sells through storytelling (editorial, lookbooks, guides, campaigns) a themed store will always feel like content bolted onto a product catalogue. A headless build with a proper CMS flips that: content and commerce become one fabric, and your marketing team gets structured, reusable content instead of fighting page builders. This is the single most common reason our clients go headless, far more often than performance.

International and multi-brand complexity

Running several markets, languages or brands from one commerce back end is where headless architecture stops being a luxury and starts being the sensible engineering answer. One codebase, many storefronts, shared logic.

Design without compromise

Themes have improved enormously, but they still have a grain, and ambitious brands feel it. Headless removes the ceiling: if your designers can imagine it, your developers can build it.

Owning your stack

Headless lets you integrate best-in-class tools (search, personalisation, subscriptions, PIM systems) on your terms rather than through whatever an app embeds into your theme. For larger organisations with existing systems, that composability is often the deciding factor.

A word on performance

Because we'd be rewriting history if we skipped it: in 2023, "headless is faster" was the headline pitch, including in our original article. It can still be true, and a well-built Hydrogen storefront is very quick indeed. But a well-built modern theme is also very quick, and a badly built headless storefront is slower than both. Speed is now a function of build quality, not architecture. Anyone still selling headless purely on speed is selling you 2021.

Who should (and shouldn't) go headless

This was the most-read section of the original post, so here it is, updated and still honest.

You're probably a good fit if:

• Content is central to how you sell, and your team publishes constantly
• You operate multiple markets, languages or brands from one back end
• You have (or will retain) genuine engineering capability, in-house or through a long-term partner
• Your design ambitions consistently outgrow what themes allow
• You need deep integrations with systems beyond Shopify's app ecosystem
• You're on Plus with the revenue to match: the platform starts at around $2,300 per month on a three-year term, and a headless build adds meaningful cost on top

You should probably stay with a theme if:

• You don't have ongoing developer access. This is the big one. With a theme, your team can change almost anything through the editor. Headless makes developers a permanent dependency for structural changes. Shopify's own guidance now says exactly this, and we agree with it.
• Your budget is tight. A serious headless build is a five-figure project at minimum, usually six for Plus-scale brands, plus ongoing maintenance. If that spend would starve your marketing budget, the maths doesn't work. Better marketing on a good theme beats a starved headless store, in our experience, far more often than not.
• Apps do heavy lifting for you. Many Shopify apps inject their functionality straight into themes. Go headless and each one becomes an integration task, or a feature you rebuild yourself. Audit your app list before you decide, not after.
• A theme already does what you need. With Horizon-generation themes, that covers more merchants than it did in 2023. If your honest answer to "what can't we do today?" is a list of nice-to-haves, a well-built theme may serve you better, even though we build headless storefronts for a living.
• You want the newest platform features on day one. Shopify ships some things, particularly in the current AI wave, to themed stores first. Headless teams sometimes wait, or build equivalents.

If you read that second list and felt seen, that's a perfectly good outcome. A fast, well-crafted theme on Shopify Plus is a strong position, not a consolation prize.

Our take

Three years on from the original post, our view has actually sharpened rather than softened: headless Shopify Plus is a better proposition than it was in 2023, for a smaller group of merchants than the industry claimed back then.

The tooling is dramatically better. Hydrogen on React Router 7 with free Oxygen hosting is a legitimately pleasant stack, and the framework-agnostic direction previewed this spring suggests the walls will keep coming down. Paired with Sanity for content, it's how we build most of our Shopify work, and the results (for the right clients) justify the investment many times over.

But "the right clients" is doing real work in that sentence. Headless is an architectural decision with a payroll attached. If content, internationalisation, design ambition or integration complexity are genuinely holding your growth back, it's one of the best investments you can make. If they aren't, keep your money and put it into product and marketing.

If you're weighing it up and want a straight answer rather than a pitch, talk to us. We've been building headless storefronts on Shopify, Centra and BigCommerce for years, and we'll happily tell you if we think a theme is the smarter call. It wouldn't be the first time, and our clients tend to trust us more for it.