Read anything about conversion rate optimisation and you’ll be told to “test everything”. That advice comes mostly from large stores with lots of traffic. For a small store, the right tool depends on one thing more than any other: how many visitors and orders you have.
Here are the three main approaches, what each one needs to work, and the order that makes sense when traffic is limited.
Conversion audits
An audit is a structured review of your store against known conversion principles: is the value clear, is there enough trust, is the path to purchase obvious, is the page fast. It can be done by a consultant, by you with a checklist, or by an AI tool.
- What it needs: no traffic at all. An audit looks at the store, not at visitor data.
- How fast: minutes with an automated tool, days or weeks with a consultant.
- What it tells you: what is likely to be a problem, based on patterns seen across many stores.
- Its limit: it’s informed opinion, not proof. It can’t tell you how your specific customers behave.
Heatmaps and session recordings
Tools like Hotjar and Microsoft Clarity (which is free) show where visitors click, how far they scroll, and let you replay individual sessions.
- What it needs: enough visits per page for patterns to appear. A heatmap based on a few dozen sessions is mostly noise.
- How fast: days to weeks of data collection, plus the time to actually watch recordings.
- What it tells you: what people do. You’ll see rage clicks, ignored buttons and where attention drops off.
- Its limit: it shows behaviour, not reasons. You still have to work out why people do what they do, and reviewing recordings takes real time.
StorePeek has a fuller comparison of the two approaches in StorePeek vs Hotjar. They answer different questions, and many stores end up using both.
A/B tests
An A/B test shows two versions of a page to different visitors and measures which one converts better. It’s the only one of the three that gives you causal evidence that a change worked.
- What it needs: a lot of traffic and, more importantly, a lot of conversions. To detect a modest improvement with confidence you typically need hundreds of orders per version. Many small stores simply can’t reach that in a reasonable time.
- How fast: weeks per test, often longer for small stores.
- What it tells you: whether a specific change made a measurable difference.
- Its limit: with too little traffic, tests either never finish or produce “winners” that are really just chance. Google Optimize, once the default free option, was shut down in 2023, so most tools are now paid.
At a glance
| Audit | Heatmaps & recordings | A/B tests | |
|---|---|---|---|
| Traffic needed | None | Moderate | High |
| Time to first result | Minutes to days | Days to weeks | Weeks or more |
| Tells you | What’s likely wrong | What visitors do | What actually works |
| Best for | Finding obvious fixes | Understanding behaviour | Validating big changes |
The order that makes sense for a small store
1. Start with an audit and fix the obvious
Most small stores have a handful of clear problems: a hidden add-to-cart button on mobile, no reviews near the price, shipping costs that only appear at checkout. You don’t need data to justify fixing those. An audit finds them in minutes, and fixing them is usually a theme setting or a small layout change. (Here are ten common ones on mobile product pages.)
2. Add a free heatmap tool to see what’s left
Once the obvious issues are gone, install a heatmap and recording tool and let it collect data for a few weeks. Look for the pages with the most traffic and the biggest drop-off, and watch a handful of sessions on those pages.
3. Test only when traffic allows, and only big changes
If you have the orders to support it, A/B test the changes you’re genuinely unsure about, like a new product page layout or a different offer. Skip testing button colours. With limited traffic, small tweaks will never produce a reliable result.
Fix what’s clearly broken first. Measure what’s left. Test only what you can’t decide.
If you want to start with step one today, StorePeek runs an AI conversion audit of your store on desktop and mobile, and the first audit is free. Freelancers and agencies auditing client stores can also use white-label reports.