ChatGPT Ads are now live in the UK. We are onboarding early advertisers. See how it works

Insights · Convert · A/B testing

A/B testing for small businesses: how to run your first test

The short answer

A/B testing shows two versions of a page or ad to different visitors at the same time and measures which brings more enquiries. For a small business, test one big change at a time, such as the headline, the offer or the form length, run each test for full weeks until you have enough enquiries to trust the result, and judge it on qualified enquiries rather than clicks.

I love A/B testing because it settles arguments. Instead of debating whether a shorter form would work better, you test it and let real visitors decide. Here is how to run your first test properly.

What is A/B testing?

Half your visitors see version A, half see version B, at the same time. You compare how many from each group enquire. Because both run together, seasonal and day-of-week effects affect both equally.

What should you test first?

TestWhy it is a good first test
HeadlineThe first thing everyone reads
Offer"Free consultation" vs "Fixed-fee quote in 1 hour"
Form lengthFewer fields often means more enquiries
Proof positionReviews beside the form vs lower down
Call to action"Book a call" vs "Get my quote"

How long should a test run?

  • Run for whole weeks, so every day of the week is included.
  • Wait until each version has a meaningful number of enquiries, not just visits.
  • Do not stop early because one version is ahead after two days.
  • Use your testing tool’s confidence figure as a guide.

What if you do not have much traffic?

With low traffic, a true A/B test can take months. Instead, make research-led changes, using heatmaps, session recordings and what your team hears on calls, and compare results before and after over the same length of time.

Which tools can you use?

Google Optimize closed in 2023. Today, tools such as VWO and Convert run website tests, many landing page builders have testing built in, and Google Ads and Meta both offer their own experiment features for ads.

Tom O, Senior Performance Specialist at Lead Gems

About the author

Tom O, Senior Performance Specialist. Tom runs campaigns across Google, Meta and LinkedIn and leads our landing page projects, making sure every pound a client spends has a clear commercial purpose.

Meet the team →

Questions

Questions, answered.

What is A/B testing?

Showing two versions of a page or ad to different visitors at the same time and measuring which brings more enquiries.

How much traffic do I need for A/B testing?

Enough for each version to collect a meaningful number of enquiries. Low-traffic sites are often better served by research-led changes.

How many things should I test at once?

One change per test, so you know what made the difference.

Can I A/B test Google Ads?

Yes. Google Ads experiments let you test ad copy, landing pages and bidding against your current setup.

Still have a question? Call 020 4652 1887 or book a free growth call.

Keep reading

More insights.

Practical guides on winning clients from search, ads, AI and automation.

Cookie settings

Choose which cookies we may use. You can change this at any time from the link at the bottom of every page.