Why we built Netaj
The big A/B testing platforms are excellent at what they do — and excellent at charging for it. Pricing tiers that lock the basics behind enterprise contracts. Editor UIs that take an hour of training. Reports that hide the actual statistics behind opinionated "winner" badges.
We wanted three things, and the existing tools made all three hard.
1. Anyone on the team can run an experiment
Not just the senior PM. The setup should take three clicks. The visual editor should feel like every other modern web tool — direct manipulation, undo, preview before launch. If a junior designer can run their first test on Tuesday afternoon and have results by Thursday, the whole company gets faster.
2. The math is honest
We show p-values and effect sizes plainly, with a clear "don't stop yet" indicator until you've hit minimum sample size. We'll never declare a winner two days in just because one variant got lucky.
3. The tracker doesn't slow your site down
Six kilobytes gzipped. Async by default. Lazy-loads the recorder only when an experiment needs it. Your Lighthouse score doesn't suffer for the data you collect.
That's it. Three things. Everything we've built since flows from those.