Mobile PageSpeed from 35 to 92: +20% conversion rate (before/after)
Test run byVastaCRO agency
Commentary byLiftArchiveOur analysis, not theirsAfter a speed overhaul took mobile Lighthouse scores from 35 to 92 on the homepage and from 30 to 91 on product pages, conversion rate rose 20%.
Vasta did the work for an anti-aging supplement brand.
This was a before-and-after comparison, not a split test.
- Conversion rate: +20% after the speed work.
- Total sales: +13%.
- The agency put the return at 21 times the cost of the work.
- These come from comparing periods before and after, so other changes over time may be part of the gain.
Control
Slow homepage and product pages on mobile. Hero images appeared after 4 to 5 seconds on 4G, third-party scripts tied up the phone's processor, and the layout shifted while loading.
Variant
The same pages after speed work, scoring 92 on the homepage and 91 on product pages in mobile Lighthouse.
Why it likely won
Our reading of the result, not the practitioner's claim.
- Visitors were likely leaving before the page had even shown the offer.
- Cutting heavy third-party scripts makes mid-range phones respond faster.
- Less layout shift means fewer mis-taps and less frustration.
Caveats: Before/after comparison, not a split test, so seasonality, ads or other changes could explain part of the lift. Time periods and traffic were not shared.
Who ran it
Vasta ran this test for an anti-aging supplement brand and published the results (Vasta case study). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Vasta →
How to run this test on your store
- Hypothesis
- If we cut load time and layout shift on mobile, then conversion rate will rise because more visitors see the offer before giving up.
- Primary metric
- Mobile conversion rate
- Guardrails
- Bounce rate
- Core Web Vitals scores
- Revenue from removed apps or scripts
- Who should test it
- Stores with low mobile speed scores and lots of third-party apps. If your site already loads fast, gains will be much smaller. Speed is hard to split test, so a holdout or staged rollout gives cleaner proof than before/after.
Questions
Does page speed affect conversion rate?
Often, especially on mobile. Here conversion rate rose 20% after scores went from 35 to 92, though it was not a split test.
What slows down a Shopify store most?
In this case, large hero images, third-party scripts and layout shift.
Apps that load scripts on every page are a common cause.
How can I test speed changes properly?
Speed is site-wide, so split testing is hard.
Comparing like-for-like periods, or rolling out to part of the traffic first, gives better evidence than a simple before/after.
Similar tests
Vasta
PrismflyPrismflyPrismfly
@kasper_dolk
Blend Commerce