Rebuild the store, or find out what is on it first
A rebuild is the most expensive answer to a question nobody has asked yet. The question is what your product page is currently loading, and it is usually answerable in an afternoon.
Fix first, and only rebuild if the fix cannot reach your number. A slow store is usually slow because of what has been added to it over the years rather than because of what it was built on, and that is a deletion job, not a project.
The theme is rarely the problem
Stores accumulate. Every app installed for a campaign that ended, every script added for a report nobody reads, every widget from a plugin the team stopped using. Each one loads on every page view, forever, until somebody counts them.
A rebuild removes all of that, which is why rebuilds appear to work. But so does removing all of that, and it costs a fraction as much and does not put your checkout at risk during peak season.
When a rebuild is actually the answer
There are real cases. A platform that cannot express your catalogue. A checkout you are contractually not allowed to touch. A theme so far from its upstream that no update can be applied. In each of those the constraint is structural, and no amount of deletion reaches it.
What those cases have in common is that you can name the constraint in one sentence. If you cannot, you are buying a rebuild to avoid an audit.
The order that costs least
Count what loads. Remove what nothing depends on. Measure again. Only then decide whether what remains is a platform problem, and by that point you will be deciding it with evidence instead of a quote.
- Count every script on a product page and name its owner
- Delete what no live feature depends on
- Rebuild in the theme only what is still earning its load
- Measure the same page again before anybody discusses platforms
What to take from this
- 01Most slow stores are slow from accumulation, not architecture
- 02Deletion is faster, cheaper and safer than replacement
- 03Rebuild when you can name the structural constraint in one sentence
- 04Never touch checkout inside five weeks of peak
Ask the question directly
- How long does the audit take?
- Counting what loads on a product page is a day. Deciding what can go safely is the slower part, because it means asking who still uses each feature.
- Will we lose functionality?
- Only the functionality nobody was using, which is what the audit is for. Anything still earning its place gets rebuilt as code inside the theme rather than deleted.
What it looked like on a real engagement
Deleting nine apps beat rebuilding the store
A Shopify storefront carrying fourteen apps, losing mobile buyers before the page finished painting.
CrUX field data, 28 day rolling window, 60 days after launch. Source: Search Console and Shopify analytics, March 2025
Related answers
Web development
New builds, rescues and re-platforms, measured against what the site earns rather than against how it looks. The brand system that carries it is built in the same engagement.
SEO or Google Ads first?
Bring the decision you are stuck on
A call, forty five minutes, no deck. We will tell you if we are the wrong firm.