Should You Migrate to Shopify's Horizon Theme?

By Solo Media Group, a Shopify Platinum Partner. Last reviewed September 2026.

What is Horizon?

Shopify's blocks-first theme family, and the current default for new stores.

• Ten presets sharing one engine. Identical framework, blocks and AI tooling; different default fonts, hero treatments, product cards and section presets. Ritual and Fabric assume hero imagery and editorial pacing. Tinker assumes grids and specifications.
• Free, with a lifetime license and one-time install. Shopify's in-house theme team ships updates weekly.
• Eight levels of nested blocks against Online Store 2.0's two. That's the headline architectural change and the part merchants most underestimate.
• About 28 base sections versus Dawn's roughly 17, plus whatever the AI block generator produces.
• Native translations in English, French, Italian, German and Spanish, without requiring Shopify Markets.
• Required for Shopify's AI theme editing. Generated layouts and content blocks only work on Horizon-based themes.

Why there's no migration path

Dawn organizes a page as sections containing blocks, two levels deep. Horizon organizes it as a block tree eight levels deep.

Those aren't versions of the same idea. They're different models, and nothing automatically translates one into the other. Shopify didn't ship a converter because a sensible one isn't possible.

What that means practically: switching to Horizon resets your home page to Horizon defaults. Custom sections don't port. Custom code doesn't port. You rebuild every template you've customized.

That's not a criticism of Horizon. It's just what the change actually is, and most merchants picture a weekend theme swap instead.

The risk nobody quotes in the estimate

Here's the argument that should weigh heaviest, and it's about conversion rather than cost.

A mature Dawn store encodes years of invisible fixes. Images given explicit dimensions so they don't shift. Fonts loaded so they don't flash and reflow. The LCP hero prioritized so it paints early. App embeds constrained so they don't inject after paint and push everything down.

Nobody documented those. They were fixed one at a time, over months, usually in response to a Core Web Vitals report.

A rebuild inherits none of them. You get a fresh block-based build and a fresh chance to reintroduce every layout-shift source you'd already eliminated. Google's CLS threshold is 0.1, and product pages and checkout are exactly where a shift costs money.

That's a different project with a different risk profile, and it belongs in the decision. More on what else breaks during a theme change in our guide to what breaks when you redesign a Shopify store.

The app compatibility problem

Horizon's variant picker, gallery and cart drawer are web components inside Shadow DOM.

That's sound architecture and it broke a decade of app convention. The standard pattern — find the variant <select>, listen for change — can't reach inside Shadow DOM. App developers have had to write Horizon-specific rendering from scratch.

Apps built for Online Store 2.0 without Horizon-specific code frequently misbehave around exactly those three elements. Review widgets, upsells, order forms, page builders, anything that injects into the theme.

When migrating is right

• You're launching a new store. No Dawn investment to protect, and Horizon is where Shopify is investing. Straightforward yes.
• You're already rebuilding. If a redesign is happening anyway, do it on Horizon. The rebuild cost is already in the budget, which removes the main objection entirely. This is the single best moment to move. Our guide to buying, customizing or building a theme covers that decision in full.
• Your current build is a mess. Years of accumulated patches, a theme three major versions behind, custom code nobody understands. You're rebuilding either way.
• You pay a developer for every custom section. Eight levels of nesting and 28 base sections mean your team can build layouts in the editor that previously needed code. If that's your recurring friction, Horizon addresses it directly.
• You run a multilingual EU store. Native EN/FR/IT/DE/ES without an app or Markets configuration.
• Layout flexibility is your brand. Fashion, home, editorial-led merchandising where the arrangement of a page is part of the product.

When to stay on Dawn

• It's customized, green on Core Web Vitals, and converting. You'd be trading a proven asset for an unproven one to gain editor flexibility. That's a bad trade for a store that's working.
• You're mobile-heavy and already scoring well. Dawn currently leads by roughly 30 PageSpeed points on mobile — an independent 2026 benchmark measured a clean Horizon install at about 52 against Dawn's 82 on identical demo content, so treat it as a directional gap rather than a figure for your store. If you're at 85+ and mobile drives your revenue, a migration is a downside risk with no guaranteed upside. See Shopify speed optimization for what actually moves those numbers.
• You depend on theme-injecting apps you haven't verified. Check first. The answer may change the decision.
• You have no bandwidth this quarter. A re-theme touches every template simultaneously. It isn't a task you fit around other work.
• You need stability above all. Horizon is young and iterating weekly. That's good for the roadmap and less good if a surprise costs you a peak trading week.

How to do it properly if you're going

1. Choose the preset deliberately. Install two or three to a development store, load your real product images, and compare on mobile. Don't pick from screenshots. Switching presets mid-project means redoing your block trees, so budget the decision now.
2. Build on an unpublished theme. Online Store → Themes → Add theme → Horizon. Nothing customer-facing changes until you publish.
3. Audit your apps first. Every app, verified for Horizon with the vendor.
4. Record a baseline. Core Web Vitals, conversion rate by device, page speed on your top templates. You need this to prove the new theme is at least as good.
5. Rebuild template by template, testing variant images, swatches, combined listings, cart drawer and checkout in detail. Those four are where Shadow DOM issues surface.
6. Measure before publishing. Compare against the baseline. If mobile CLS or LCP regressed, fix it before launch rather than after.
7. Consider Theme Rollouts. Shopify now lets you publish a theme as a scheduled release or run it as an A/B test against your current one. For a migration this valuable, testing rather than switching is the obvious move.
8. Keep Dawn in your theme library for at least 30 days. It's your rollback and your reference.

A note on tooling

Worth knowing if your team writes Liquid: Shopify's Dev MCP server runs locally and exposes Shopify's developer docs, GraphQL schema and theme architecture to MCP-compatible AI clients.

Connected to Claude Code alongside the Shopify CLI, it can read your Dawn sections/ folder, map each section to a Horizon block structure, and generate the migration diff with current syntax rather than stale training data. It can also run shopify theme check, parse the output and propose fixes.

Treat Theme Check errors as blockers. Wired into a pre-commit hook or CI step, nothing untested reaches your Horizon theme.

This doesn't make the rebuild automatic. It makes the mechanical half considerably faster, which is a genuine change from six months ago.

Frequently asked questions

Weighing a Horizon migration?

Tell us your current theme, your mobile Core Web Vitals, and which apps inject into your storefront. If your Dawn store is converting and passing, we'll tell you to stay — that answer costs you nothing.

Book a Discovery Call