Ask HN: How to convince my company to avoid dark patterns? | Hacker News Reader

Ask HN: How to convince my company to avoid dark patterns?

3 pointsgingerlime1 comment

I'm a co-founder of a small bootstrapped company. I also created an open source A/B testing framework[0] that we use to run A/B tests at my company. It helps us answer certain hypotheses and improve the user experience, as well as conversion.

Lately we've ran an experiment that felt to me like it crossed the line into dark pattern territory. It offered bonus eBooks for those who purchase within 24 hours on the pricing and checkout pages. With a ticking clock to build urgency... textbook example of a dark pattern?[1]

Despite being a co-founder and leading A/B testing at my company, I'm not the only one calling the shots, and others were keen on running this experiment. In their mind it was simply a nice gift that we offer to customers. They couldn't see the potential harm.

Luckily(?) this experiment didn't fare that well, so we're in the clear. For now. I'm wondering how to make them see the dangers and temptation with dark patterns, especially if they increase conversions. Any thoughts / advice?

[0] https://github.com/Alephbet/Alephbet

[1] https://webtransparency.cs.princeton.edu/dark-patterns/

Only as a side note, and of course IMHO, A/B testing may not necessarily - even if sometimes it happens - end up with actually improving user experience.

What it does - once a result is chosen - is to "shift" user experience towards what an unqualified numerical majority likes more (which not always is the same as better), and while the experiment is running it risks to irk a lot of users.

To the scope of making the whatever more popular. it is of course appropriate, but it shouldn't be called "better" or "improvement" unless it is actually better or it really is an improvement.

About the specific on dark patterns, you must I believe, re-consider your (or your firm's) approach, as you say "luckily " it didn't work this time, but that doesn't mean that the next dark pattern will not work, possibly surprisingly well.

Dark patterns exist exactly because they are successful (in sheer numbers of subscriptions, conversions, etc.), even if often only on the short or very short term.

We cannot deny this.

Many companies use dark patterns and are successful until they cross an invisible line that - all of a sudden - makes their customers angry at them.

In a perfect world customers at the very first fraction of a hint that maybe a dark pattern is used should get angry and leave.

But this is not what happens in real life, they come and stay until they don't.

So you won't be able to convince anyone that dark patterns are "bad" (as in the short term they do work) if not by scaring them about the possible long term effects or by making ethical considerations (historically things like a company motto "Don't do evil" didn't work all that well)

Reply on news.ycombinator.com