4 ms·
It's funny how smart yet dumb A/B testing is. On one hand you can intelligently gauge the effect of changes, on the other hand you can push stupid shit since yo
by 2OEH8eoCRo0 4y ago
It's funny how smart yet dumb A/B testing is. On one hand you can intelligently gauge the effect of changes, on the other hand you can push stupid shit since you have that power.
How about intelligently designing applications that you yourself want to use? Too hard.
- eurasiantiger 4y agoThat would require designers to deeply understand system and software engineering, or systems and software engineers to understand design. Funny how that gap keeps on manifesting itself, even though both parties work on the same domain.
- dtgriscom 4y agoThere are facts about human behavior which can only be learned by actually testing them. People are complex in ways that you, whilst sitting at your keyboard, cannot figure out.
- tornato7 4y agoAll which can be learned through focus groups and internal feedback, no need to further confuse and frustrate your users.
- szundi 4y agoAfter you have the engine running, the marginal cost of an A/B is probably smaller than a focus group test. So if you really believe in focus groups for the results and you are also ambitious enough - probably this means go for A/B. Disclaimer: just thinking, not knowing the costs
- tornato7 4y agoDefinitely agree, though I think the point of this thread is we don’t know the “true cost” of A/B testing in terms of long-term lost user engagement. Have we done an A/B test for A/B testing where we test A/B testing…
- fshbbdssbbgdd 4y agoIt’s very common to think something is a good idea based on focus groups and internal feedback and then learn it’s a bad idea from launching it (or running an AB test).
- collaborative 4y agoAll the A/B tests I have done point to one conclusion: assume users are illiterate and have an attention span of 5 seconds
- eurasiantiger 4y agoKeep it simple, stupid.
- notriddle 4y ago> How about intelligently designing applications that you yourself want to use? The people who use the apps I make aren't always me. Many of them are very good at things that I know nothing about, but I usually know more about computers than they do. Sometimes the differences are even harder to account for, having different preferences, different ways of thinking, and participating in different cultures.