Typically you'll run 8~10 A/B tests to get 2 significant results of which one will be significant in a happy direction.
-
-
Replying to @patio11
This counsels making tests cheap/free operationally, having no political barriers to doing them, and not getting attached to winning.
1 reply 1 retweet 7 likes -
Replying to @patio11
You're running a portfolio strategy. Losers: expected. Your best idea being indistinguishable from the intern's off-the-cuff idea: expected.
1 reply 2 retweets 11 likes -
Replying to @patio11
Most people are not wired to make lots of little bets and lose again and again and again to win big. Communicate to help them understand.
1 reply 6 retweets 17 likes
Typically this means batching reporting of results on a defined cadence (so folks know to expect it) and say "Learned X, Y, Z. Biz impact:"
0 replies
1 retweet
12 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.