Hacker News Reader

Comment by gavribirnbaum | Hacker News Reader

I loveee the concept of drawing rooms with a mouse to enable spontaneous conversations. the site is incredibly wordy and doesn't drive a CTA, but that is really nice. when can I use it?

View on HN

No everything on the internet needs to be about pixel trackers, A/B testing, micromanipulating users, conversion rates, ARPU, analytics and all that boohoo. Can we just pause for a second and be honest with our users? May be respect them? Try to make insanely great products that people have no other option than to fall in love with them?

In fact, I want to go to sites that don't lie to me and backstab me, treat me like some kinda of a ephemeral datapoint consuming their giant wad of VC cash.

I call these people sales-assholes or salesholes.

A/B testing is in fact often used to make insanely great product, how would you otherwise know what your users actually want

> how would you otherwise know what your users actually want

E.g. with regular user testing.

Probably not applicable everywhere, but could replace a/b testing for some products.

Perhaps understanding the problem deeply, thoroughly and putting yourself in the shoes of the customer insteading throwing a bunch of semi-random iterations over the fence and squeezing every last bit of data from the customer's private behavior?

Imagine if I buy tomatoes from the grocery store, but they run an A/B testing on me to see what the customer wants - may be I prefer the 85% ripe ones over 95% ripe ones, the farmers watch the reaction of their users as they cook and taste tomatoes in a live video stream from their customer's home. WTF, get out of my house (browser). It is my private property. I just want a goddamn tomato (send an email or post a photo or read news).

What happened to user's privacy? That last mile optimization is bothering me as a user.

In many cases A/B testing, conversion rates, analytics are used to actually improve user experience and create a better product. I don't think the problem is the analytics themselves, but what you do with the data. For example, you can see based on analytics that users spend more if feature A is implemented but they have a better experience and rate product higher if feature B is added instead. The stats allow you to have this insight, but then it's up to the company if they want to go for profits only or to actually create a better product.

Reply on news.ycombinator.com