Hacker News Reader

Comment by joshuacc | Hacker News Reader

The way that they preselect an amount $1 above average is extremely clever. It appeals to people's need to see themselves as above average without costing them anything substantial.

Screenshot: http://screensnapr.com/v/QciSbr.png

View on HN

I'd love to be a fly on the wall for the A/B test, but I still think something a wee closer to

http://images1.bingocardcreator.com/blog-images/hn/deep-disc...

would kill both the A/B tested options. Bringing in people's psychological need to be above average is great, but I might do that by putting a nice 37Signals style yellow highlight on one of the expensive options (for some value of expensive which would pay for more than a Big Mac) and say "24% of people went for this"

I'd love to see your variation tested too but the data-driven "above average" feels much less accusatory and supportive than subjective labeling of "poor student" and "valued contributor". Makes me wonder if poor students are less valued?

Also the fine-grained payment distribution scrollbar controls make me feel empowered and better about my participation in this cause. Getting rid of them would disappoint me.

This reminds me of someone's donation button on some open source project, it was something like:

    Donate [$5 — buy me a beer] [$20 — buy me dinner] [$100 — buy me a gadget]

Personally, "Tired of overpriced games from soul-crushing megacorps laden with spyware and DRM?" really puts me off.

Are you tired of underpriced games from small indie dev shops without spyware or DRM?

It's a somewhat antagonistic statement, and those can rankle.

ooh nice one!

Hope you don't mind me using that mockup for one of my own experiments :)

Go for it. I'd love to know if it works out for you.

If you don't mind me asking, what did you use to make that mockup image?

I Believe it's Balsamic Mockups http://balsamiq.com/products/mockups

and looks like they're a/b testing it; http://i.imgur.com/NsjwG.png

A/B/C testing: http://goo.gl/68ZXm

I'd be curious to know which version was the most effective in the end.

Yep. The JS looks like they are doing it with Visual Website Optimizer.

And it will gradually increase the average every time someone chooses that option.

I thought the same when I saw it.

Weirdly enough, though, the average was at about $4.40 when I saw it earlier today, and it is $4.27 now.

Wonder if this technique might be driving some people to pay more (I will do it as soon as I get home), but not enough to offset all the people that just want lots of games for no money at all.

There is also the possibility that it is cannibalizing their higher-paying customer base. For instance, perhaps if I hadn't been presented with that option I would have paid $10 instead. In that case it would drive the average down rather than up.

Presumably that's why they're A/B testing it. :-)

If the average is currently $X after N purchases, that means they have taken in $(N * X). When the next person pays $(X + 1), the total is now $((N * X) + (X + 1)) == $(((N + 1) * X) + 1). The new average is $(X + (1 / (N + 1))), which is guaranteed to be higher than the previous average, even if $(X + 1) < $10.

Yes, the average after purchase (AAP) would be greater than the average before purchase (ABP). However, the AAP of $10 is greater than the AAP of $X+1.

The question is to what extent the $X+1 purchases depress greater purchases, as well as to what extent the $X+1 purchases increase smaller purchases.

The answer can only be determined by actually testing it.

Oh, I totally would have fell for that!

I got the scenario that janzer posted, and I opted for $5 since I wasn't too excited about any of the games and just wanted to support the cause. I felt a little cheap because the cheapest example was $10, but felt justified based on the selection of games. If $5.81 was the default option with "above average" label, I would have went with that.

Indeed, and I went for it in a shot - worked very well.

Reply on news.ycombinator.com