About Experimento

Experimento is the field journal for conversion and product experimentation. We publish A/B teardowns, the maths behind statistical significance, guides to the tools people actually run tests with, and the playbooks that move a funnel.

It is for the people doing the work: growth and product teams, CRO specialists, designers, and founders who would rather check an assumption than argue about it. We are independent. No vendor pays for coverage, and no tool buys its way into a recommendation.

How we research

We start with the test. Where we can, we run experiments ourselves or rebuild a documented one to see whether the result holds, then show the setup, the sample size, and how we read the numbers so you can disagree with our working.

For statistics we follow established methods rather than rules of thumb: standard frequentist significance testing, sample size and power calculations, and sequential or Bayesian approaches where they fit the question. We cite the source behind a claim, whether that is a primary study, a tool’s own documentation, or published guidance from people who have written the textbooks on experimentation. Figures and tool details are dated, because pricing, features, and platform behaviour change.

We do not invent named experts or attach credentials to people who do not exist. Our work is produced by the Experimento editorial team. When we lean on outside research, we name the source and link to it so you can read it yourself.

Editorial standards

We aim to be accurate and specific. We prefer a worked example over a slogan, and we would rather say “we are not sure” than overstate a result. When we get something wrong, we fix it and note what changed, with the date. If you spot an error, tell us and we will correct it.

The writing is meant to be plain and useful. If a concept needs a formula, we show the formula; if it needs a caveat, we keep the caveat.

A note on links

Some outbound links may be affiliate links, which means we may earn a commission if you buy through them. This never affects our recommendations. We rate a testing tool the same way whether or not there is a link, and we will tell you when something is not worth paying for.