-
I have a question: purely from a business perspective, is there anything that has actually proved that algorithmic recommendations are better for retention than chronological? Like a study? kevinroose/1137372560113274881On twitter.com
♻️ 11 Retweets
❤️ 41 Favorites
Mood +4 🙂
-
Like, we just take the assumption that algorithmic recommendation is needed for these companies and sites to work... but all of these platforms did their peak onboarding when they were using mostly chronological models.Permalink On twitter.com
♻️ 2 Retweets
❤️ 25 Favorites
Mood +2 🙂
-
The thing about a/b testing is that if you test two wrong answers you'll still get a "right" answer that also happens to be wrong.Permalink On twitter.com
♻️ 8 Retweets
❤️ 35 Favorites
Mood -4 🙁
-
If there is one thing I learned definitively from working with ad tech it is this: don't trust that anyone, even assumed experts, know wtf they are doing.Permalink On twitter.com
♻️ 2 Retweets
❤️ 17 Favorites
Mood -5 🙁