undefined | Better HN

0 pointsraw_anon_111111d ago0 comments

Would you rather they change things for everyone at once without testing?

0 comments

That is not the only other alternative.

You can do A/B testing splitting up your audience in groups, having some audience use A, and others use B - all the time.

I think the article’s author is frustrated over sometimes getting A and at other times B, and not knowing when he is on either.

j / k navigate · click thread line to collapse

0 pointsraw_anon_111111d ago0 comments

Would you rather they change things for everyone at once without testing?

aeinbu11d ago

That is not the only other alternative.

You can do A/B testing splitting up your audience in groups, having some audience use A, and others use B - all the time.

I think the article’s author is frustrated over sometimes getting A and at other times B, and not knowing when he is on either.

j / k navigate · click thread line to collapse