The book refers to ‘inconclusive results’ starting on page 12. I would caution against introducing this term without carefully defining it. At my dayjob I frequently deal with product managers who believe that an A/B experiment with a neutral outcome is ‘inconclusive’, when in fact it just says that there is no statistically significant difference between the two arms.
That’s a really good point. I’ll do exactly as you suggest, define and give caveats. Luckily we’re still in beta so I can make these changes this week. Thank you David!
That’s a really good point. I’ll do exactly as you suggest, the details matter. I’ll define and give caveats of what it looks like in practice/what that could mean. Luckily we’re still in beta so I can make these changes this week. Thank you @dlindelof!
1 Like