Horizon is the biggest change to Shopify theming since Online Store 2.0, and most of what it adds is good. You get in-canvas text editing, hover-to-preview blocks, copy and paste across templates, conditional settings that hide irrelevant fields, and AI block generation from a written prompt. The editor finally behaves like a modern design tool. The feature list is the easy part of this decision.
The harder part is what it costs to get there from the theme you run today, and the honest answer is that there is no migration path. Shopify did not build an automatic converter because there is nothing to convert. Horizon's block tree is a different structural model from Dawn's section architecture. Moving means rebuilding every customised template, section by section, then re-validating every app integration against the new structure.
What breaks, specifically
- ✓Custom sections and template variations do not carry over. You rebuild them by hand.
- ✓Apps that inject blocks or markup against Dawn-era layout assumptions can render incorrectly, or not at all, on Horizon's restructured output
- ✓Merchants report media and aspect-ratio inconsistencies on early Horizon versions, especially on image-heavy product and collection layouts
- ✓Years of small theme customisations (badges, swatches, bundle blocks, shipping bars) each need an explicit decision: rebuild, replace, or drop
Timing matters more than the feature list
It is August. A Horizon migration on a customised store is a proper rebuild project, and rebuild projects compress badly. Start one now and you will be QA-ing a new theme architecture in the same weeks you should be locking the store down for BFCM. Unless your current theme is actively failing, put Horizon in the January to March window. A rebuild can breathe there, and a problem costs you a quiet Tuesday instead of peak-season revenue.
"A theme migration is a rebuild wearing a friendlier name. Price it, schedule it, and QA it like one."
— Thought Bulb
Who should move now, and who should not
- ✓Launching a new store or a new market storefront: start on Horizon, there is no legacy to translate
- ✓Running a lightly customised Dawn-family theme: a post-holiday migration is realistic and worth planning now
- ✓Running a heavily customised theme with deep app integrations: treat it as a redesign decision with a real budget attached
- ✓Already mid-migration: freeze the scope and get one template pipeline fully working before touching the rest
Do the boring version first
In most cases, a store considering Horizon is suffering from fixable problems in the theme it already runs: slow sections, brittle app leftovers, template drift. Fixing those usually costs a fraction of a rebuild and improves the store this quarter. When you do migrate, you are moving a clean codebase instead of paying someone to faithfully recreate your technical debt on a new architecture.
Not sure whether this is a refresh, a redesign, or a rebuild? We wrote a decision framework for exactly that call.
Read the redesign decision guide →Thought Bulb scopes Horizon migrations honestly, including telling you to wait when waiting is the right answer, and handles the pre-migration cleanup that makes the eventual move cheaper.
Scope your Horizon move →
