How to Test Your Shopify Store Speed (and Read the Results Correctly)
Run the same speed test twice and you will get two different numbers. That is not a bug, and it is the main reason merchants lose confidence in speed work entirely.
The Short Answer
Test three page types — homepage, your busiest collection, your busiest product page — on mobile, and trust field data over lab scores. Everything below is detail on those three decisions.
The Tools and What Each Tells You
| Tool | Data type | Best for |
|---|---|---|
| Shopify web performance report | Field | Your real visitors, over time |
| PageSpeed Insights | Both | Field verdict plus lab diagnosis |
| Search Console | Field | Which templates fail at scale |
| Lighthouse (DevTools) | Lab | Debugging a specific page |
| WebPageTest | Lab | Controlled comparisons, blocking domains |
| FixMyStore Speed Checker | Lab | Quick Shopify-specific check |
Lab vs Field Data, in Plain English
Lab data is one simulated visit on a throttled connection. It is reproducible and good for diagnosis, but it is not your customers.
Field data is aggregated from real visits — real devices, real networks, real locations. It is what Core Web Vitals assessment uses, and it is the number that matters. It also lags, because it is collected over a rolling window.
When they disagree, the field data is right about your customers and the lab data is right about what to fix.
A Repeatable Test Protocol
- Pick the three page types and fix the exact URLs
- Test on mobile
- Run each three times and take the median — a single run is noise
- Record the results in a dated sheet
- Re-test the same URLs the same way after any change
Without a baseline sheet you cannot tell improvement from variance.
Reading PageSpeed Insights
The pass/fail assessment at the top is field data and is the verdict. The 0–100 score below is lab and is a diagnostic aid, not a KPI. Chasing the score while the assessment still fails is the most common way to waste a week.
Work the opportunities list top-down, and treat estimated savings as indicative.
Why Is My Shopify Store Slow? Six Usual Suspects
- Third-party apps and scripts — app bloat
- An oversized or lazy-loaded hero image — image optimization
- A heavy theme with many homepage sections
- Web fonts blocking text rendering
- Autoplay video
- Marketing pixels, each adding a connection
When to Re-test
After every app install, theme update and major campaign. Performance regressions almost always arrive with a change someone forgot to measure.
Frequently Asked Questions
What's a good Shopify speed score?
The wrong target. Aim to pass the field-data assessment; the lab score will follow.
Why does my score change every run?
Lab tests vary with network conditions and test-server load. Take the median of three.
Why is mobile so much lower than desktop?
Mobile tests simulate slower processors and networks. That simulation is closer to your real audience than the desktop run.
Can a low score still pass Core Web Vitals?
Yes, routinely — the score is lab, the assessment is field.
How often should I test?
Monthly as a baseline, plus after every significant change.
Where ConversionAB Fits
If your tests point at theme code or apps, ConversionAB can fix what the tools find.