Conversion Intelligence
A/B Testing on a Small Site's Traffic
Most A/B testing advice is written for sites with millions of visitors, and it quietly tells everyone else the truth is "you can't test." That's wrong. You can test meaningfully on modest traffic — you just have to test differently. The rules change, and once you know how they change, testing becomes one of the most honest tools a small site has.
The Constraint
The problem is time, not possibility
The reason big-traffic sites test constantly is simple: they reach a confident answer in days. On a small site, the same test needs longer to collect enough visits to trust — and the longer a test runs, the more the world changes underneath it. That's the real constraint: not that you can't test, but that a tiny improvement will take so long to prove that it isn't worth chasing. Which points straight at how to test well at small scale.
How To Test Anyway
Bigger swings, fewer tests, clearer wins
- Test big changes, not tweaks. A button-colour test needs enormous traffic to prove a tiny effect. Rewriting a headline to answer the buyer's actual question can produce a swing large enough to see on the traffic you have. Small sites should test the boldest honest idea, not the safest small one.
- Test higher up the funnel. The step more people reach gathers data faster. Testing the checkout when only a trickle reaches it is hopeless; testing the page everyone lands on is not.
- Run one test at a time. With limited traffic, splitting it across competing tests means none of them ever reaches a trustworthy answer. Queue them.
- Let it run its full course. The cardinal sin on small traffic is peeking early and calling a winner from a handful of conversions. An early lead on thin data is mostly luck. Decide the finish line before you start, and hold to it.
The Honest Part
Sometimes the right move is to just decide
Not everything deserves a test. If a page is clearly broken, or the copy plainly doesn't answer the buyer's question, you don't need weeks of data to justify fixing it — you need to fix it and move on. Testing is for genuine uncertainty between two reasonable options, not for outsourcing an obvious call to a spreadsheet. On a small site especially, judgement is cheap and traffic is precious; spend the traffic only where a test can actually settle a real question.
What ARIA Shows You
What's worth testing — and what to just fix
ARIA reads your funnel and points you at the tests worth your limited traffic, and the fixes that don't need a test at all. A read might come back as:
"Your homepage gets the traffic to support one real test, and the highest-value one is the headline — the copy-value read says it isn't answering the visitor's question, and that's a big enough change to show up in your numbers. Meanwhile, your contact form is broken on mobile: don't test that, just fix it. Spend the test on the headline; the form is a decision, not an experiment."
That's the small-site discipline in one read: test the bold thing where you have the traffic, and just fix the obvious thing everywhere else.
Keep Reading
More in the Conversion Intelligence guide
- Reading a Funnel Drop-Off (Hypothesis-First) — where the hypothesis a test is meant to prove comes from.
- The Copy-Value Score, Explained — often the source of the boldest, most testable change.
Find the test worth running
Connect your site and let ARIA read your funnel for the one test your traffic can actually settle — and the obvious fixes you shouldn't waste a test on at all.