Interpreting the results of Popify.me HN launch

My name is R. Yesterday we let popify.me loose into the wild and here is a follow-up on how it survived through its first night.

The Story

This gets pretty Flintstonish sometimes, so beware:

Yesterday, we finished the most basic version of popify.me. We thought that HN would be a nice place to tell people about the service, so we did.

Show HN: measure curiosity of your website users

First, we tried to find a headline that would stick at least an hour on new section, for some high karma people to see it and upvote it a bit.

But here comes the first Flintstone, me, trying to post news with a dead HN account! I didn’t even know it’s dead and I’m a daily HN reader for about 170 days now. I was just ok with my karma of 1 and didn’t read the rules of the game, because I see myself as a pretty polite person everywhere I move. I could blame it on pg, because I remember posting a comment twice in a row one day, as HN get pretty unresponsive sometimes, but I won’t, because he’s a superhuman anyways and I should know the rules if I’m playing the game.

So I asked Miha to clone the dead post and we were back in the game. We also asked a favor from Sahil of GumRoad, Franz of Facesearch & Swizec of HipsterVision to upvote our post. A nice debate also fired in the discuss section. Some questions started to appear and, guess what: Miha set his noprocrast to like 4 hours, so he was locked out of HN and couldn’t respond to questions.

Matevz came to rescue and Rok and Robi also poked around a bit with a large trout.

So, with a nice discussion going on, we even managed to land on the first page of HN and stayed there for like 6 hours. It was a nice adrenaline rush for the whole team and we learned a lot.

There were around 2000 unique visitors from HN and we got some positive Twitter coverage.

Feedback

We had these feedback tools at our hand:

  • GetClicky for realtime tracking of the visitors (we liked the Spy World view and dots popping up in Tasmania)
  • Olark for live chat support on edit page
  • A simple feedback box in Edit Coupon page and
  • The invaluable HN discuss section

Lessons

  • We definitely need a native English speaker for explainer video.
  • We won’t use Vimeo for explainers anymore because it’s blank on iOS devices.
  • English proofreading is necessary “less then 1M” should be “less than 1M” & such
  • Nudity on a sample coupon is not accepted, cows are.

And, the most important was that we soon realized, that visitors didn’t understand why would they use popify.me, which analytics clearly showed.

The analytics and A/B Testing Results

Our goal was to try to get as many people as possible to try out our coupons and spread them on Twitter/FB and, of course, put them on their web sites. Or at least try to create some. But we also wanted to learn if people were prepared to pay for such a service.

We used

Results

This was the first page:

Variation 1

We had a variation that tried to communicate extraordinary results we were getting in our test runs on some news publishing websites. If you are a geek of any sort, if I say you’ll get 3.5% of CTR or a 8% e-mail newsletter subscription rate, you should be standing and screaming already. We didn’t lie about the results, but it didn’t make such a big effect as we would think. Less than 2% of visitors engaged. But there was a price tag, which could you basically reduce or remove by just retweeting coupons. It didn’t matter. A price tag is a price tag.

Variation 2

This variation explained for every target group:

  • business,
  • webmaster,
  • party animal,

for what they should use popify.me. But it was 1 click away and only 1% of visitors clicked on it.

Variation 3

We dumped the big sign about juicy coupons and told the story behind popify.me. Out of the variations with a with a price tag, this one was the most successful.

Variation 4

This one was basically the same as Variation 1, but with no price tag. It had a 6% engagement rate. The highest of all.

Conclusion & Your suggestions

We expected a lot more engagement from 2000 users. There were around 50 coupons created and 5 embedded on the web sites. We got a lot of LIVE support and HN discuss questions about possible use cases for popify.me, so obviously we didn’t communicate this good enough.

The price tag was also a big limitation. We’ll drop that, because we don’t have any expense with the site as we are using HipHop standalone server and Redis and can serve a gazillions of coupons with a load of 0.1.

The story reflected good and it seemed that if the visitors get the point they are more likely to engage more. We’ll of course have to work more on our story-telling abilities, but that’s fun stuff anyway.

What would be your thoughts on how to introduce and explain popify.me for people to pick it up faster? Please leave a comment.

Plan

Since we think that telling the story is the best way to go, we’ll loose the big juicy sign and make 4-5 stories for different verticals to see which ones users like the most. We still have a constant influx of visitors from other sources than HN and I hope that this follow-up generates some more buzz. I’ll follow-up :)

Tags: popifyme popify hackernews ycombinator coupons