Shopify Speed Optimization

A full audit with a prioritised, written action plan is a fixed fee starting around US$500. Implementation is quoted from that audit and depends on how customised the theme is — a clean Dawn-based theme is far cheaper than a heavily modified legacy one. Ad-hoc work is US$60 to US$90 per hour, and you always approve the number in advance.

On the large majority of stores I audit it is not Shopify — it is what was added on top. The usual causes, in order of how often they turn up: a stack of apps each injecting its own render-blocking script into the head, two or three copies of the same analytics or consent tag, an unoptimised hero image with no priority hint or explicit dimensions, a slider or review widget that shifts layout after paint, fonts loaded without a display strategy, and Liquid loops that hit the same collection repeatedly on a page that could cache it. Shopify's own infrastructure is fast; almost all of the latency in a slow store is added after the theme is installed.

I am in São Paulo, UTC-3, one to two hours ahead of US Eastern, so a normal US business day leaves four to six hours of live overlap. Publishing happens in your store's lowest-traffic window, which usually means late night your time — that is my morning, so I am awake and watching when it goes live.

Yes, on every engagement, and I will use your NDA or MSA if you prefer. I access the store through a staff account you create with the permissions the work requires, I always duplicate the theme before touching anything, and I keep a rollback point so the previous version can be restored in under a minute.

You do, in full, on payment. That includes the refactored theme, any custom scripts, the audit document and the handover notes. I do not install a proprietary loader, I do not charge a monthly fee to keep the improvements working, and any other developer can read and maintain what I leave behind.

No, and be careful with anyone who does. On Shopify the platform, your apps and your marketing pixels all contribute to the score, and hitting a specific number often means breaking things your business needs. I commit to measurable improvements in real-user Core Web Vitals and to telling you exactly what a higher score would cost you in functionality.

Rarely. Most of them defer scripts indiscriminately, which improves the lab score while making the page feel worse and occasionally breaking add-to-cart. They also add another script to a store that already has too many. Fixing the theme and removing dead apps produces a better result that does not require a subscription to keep.