It wasn't a fluke -- variation was statistically significant better than control (we never publish results otherwise). Clients are never happy to see their actu

It wasn't a fluke -- variation was statistically significant better than control (we never publish results otherwise). Clients are never happy to see their actual conversion rates and variance, etc. published in the real world, that's why I choose not to write about those figures.