Service
A/B testing programme
A steady rhythm of A/B tests, designed and analysed properly. Every result, win or loss, teaches you something you can use.
Book a free discovery callWho it's for
Teams who want a steady cadence of tests, run properly — not just "try stuff and see."
What's included
- Monthly test roadmap
- Hypothesis and test design
- QA and statistical analysis
- Reporting and documentation of learnings
How it works
An ongoing monthly engagement. We agree a cadence, I run or guide the testing process, and every result — win, loss, or inconclusive — gets documented.
Step by step
- 01
Roadmap
We collect test ideas from data and research, then score them (ICE or PIE) so the best ideas go first.
- 02
Test design
Each test gets a clear hypothesis, a primary metric, and a sample size and duration calculated before launch.
- 03
Build and QA
The variant is built and checked on all key devices and browsers before any traffic sees it.
- 04
Analysis and learnings
Results are analysed with the right statistics and written up in plain language, including tests that lose.
Questions
Frequentist or Bayesian?
Either. I use the approach that fits your traffic and how your team makes decisions, and explain results in plain language either way.
How much traffic do I need for A/B testing?
It depends on your conversion rate and the size of change you want to detect. I calculate this before every test, so you never run a test that can't give a clear answer.
Do you build the tests yourself?
I can run the whole process or guide your team through it. We agree the split at the start.