Postcard Panda May 2016 Income Report

New to this series? Checkout the first report for background info on Postcard Panda.

Overall Stats

 April 2016
May 2016% Change
Total Expenses-$227.06$-462.59103.73%

Expenses Breakdown

 April 2016May 2016% Change$160$348.75117.97%
Hosting (Heroku)$7.00$7.000.00%
S3 + Cloudfront$7.55$4.39-41.85%

USA App Store Stats

 April 2016May 2016% Change
Guest Orders213566.67%
Logged In Orders100242142.00%
Total Orders121277128.93%
Signups who placed an order397079.49%

UK App Store Stats

 April 2016May 2016% Change
Guest Orders10-100.00%
Logged In Orders62-66.67%
Total Orders72-71.43%
Signups who placed an order110.00%

Cohort Analysis

As I mentioned last month, of the 30 users who originally signed up and ordered a postcard in March of 2016, 13 of them also ordered a postcard in April. Of those 30, 2 of them also ordered a postcard in May.

Of the 39 users who signed up and ordered a postcard in April of 2016, 14 of them also ordered a postcard in May.

I haven’t figured out the best way to visually represent this data yet, so bear with me in paragraph form until I figure that out.

What Went Right

  • My definition of a power user was shattered this month, by one user who ordered 71 postcards in the month of April. Since the creation of that users account, they’ve ordered 82 total postcards using Postcard Panda. Makes me wonder if she’s a “whale” customer, or if there are still bigger power users out there.
  • The reengagement numbers for the April cohort were healthy.
  • I put out two small updates to the app to fix some bugs and polish it a bit. The use contact from address book functionality has now been improved considerably.
  • I had a one day spike of 242 downloads, and an all time single day high in revenue at $71.64 on May 24th. Unfortunately, I have no clue where those downloads came from which is frustrating.
  • I launched my first a/b test with Mixpanel (the results are inconclusive at the moment).

What Went Wrong

  • The UK app store numbers slid closer to zero than it already was.
  • The March cohort didn’t do well in May. Because Postcard Panda is generally used for a particular circumstance like a vacation, it’s possible that this is normal behavior. It does point out the fact that I don’t have a reengagement strategy yet. Do I need to remind people the app exists? Or is it simple enough that people will remember it the next time they go on vacation? What’s the balance between being useful, without being spammy?
  • I had a bug inside the iPad version that made it hard to sign up in some cases, which has now been fixed. My users are heavily skewed towards iPhone so hopefully the impact was small.

Next Month’s Goals

I still haven’t finished adding Postcard Panda’s logo to the back of each printed postcard like I had mentioned at the end of last month’s report. Turns out it’s a big task which involves ImageMagick, which is always an equal parts gratifying and horrifyingly difficult tool to get just right. This task is still at the top of my list.

On the marketing side of things, I want to come up with a reengagement strategy and perhaps dabble in a paid advertising channel again. I also want to switch out the screenshots in iTunes Connect to actually showcase the printed products; instead of showing a random stock photo.


The growth percentage for May was awesome! Numbers are still low on an absolute scale, but May represents the best month in absolute numbers and relative growth that Postcard Panda has ever had. I’m a bit nervous that my power user might have skewed May higher than it should have been, but that’s a good problem to have.

Thanks for reading, and if you want to receive future updates via email please fill out the form below!

Leave a Reply

Leave a Reply

Your email address will not be published. Required fields are marked *