Postcard Panda October 2016 Income Report

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

tl;dr – Revenue was down again. I had a bug which cost me about $20 in profit, but if you don’t count that, profit was actually up compared to September, because I cut out an app store optimization tool I was using.

Overall Stats

 September 2016October 2016% Change
Revenue$420.71$307.76-26.85%
Total Expenses$394.68$269.94-31.61%
Profit$26.03$37.8245.29%

Expenses Breakdown

 September 2016October 2016% Change
Kite.ly$261.25$190.00-27.27%
Hosting (Heroku)$14.00$14.000%
Advertising$33.000-100%
Stripe$75.34$54.83-27.22%
S3 + Cloudfront$5.12$5.140.39%
Refunds$5.97$5.970%

USA App Store Stats

 September 2016October 2016% Change
Guest Orders2218
Logged In Orders177123-18.18%
Total Orders199141-29.15%
Signups who placed an order2322-4.35%
Downloads8511231.76%
Reviews110%

UK App Store Stats

 September 2016October 2016% Change
Guest Orders000%
Logged In Orders101110%
Total Orders101110%
Signups who placed an order12100%
Downloads51-80.005
Reviews000%

Retention

screen-shot-2016-11-11-at-11-07-52-am

What Went Right

  • With the exception of the bug fix that I’ll talk more about below, the app was largely on auto pilot. It’s nice to let it run without having to invest a ton of time into it.
  • Profit was up nicely because I dropped my app store SEO tool. I actually liked the tool, but after 3 months of using it, I wasn’t seeing results.

What Went Wrong

  • I had a bug in the back image creation code I pushed up last month. Specifically, if the message was short and didn’t wrap to two lines the code didn’t actually write a message. I had to reprint 22 postcards out of my own wallet after fixing the bug, which lost me an extra $22.50 that I didn’t include above.
  • The above bug resulted from sloppy coding on my part, without proper unit tests in place. I rarely write code without unit tests and fell into the ‘prototyping trap’ of many organizations. I had prototype code that appeared to work, but was poorly written. Instead of going back and re writing the code, I just called it good enough and pushed it to production. Note to future Matt… write tests for all the things!

Next Month’s Goals

The top priority for next month is to try a new marketing channel for Postcard Panda, which will likely be Apple’s new search ads.

Thanks for reading, and check back next month!

2 Comments

·

Leave a Reply

  1. Catching up on these reports since you haven’t been tweeted about them lately 🙂

    Bummer about that bug, but it sounds like a good lesson came out of it! And, as always, it’s great to hear about you doing the right thing for your customers (reprinting all of them—it may seem like a no-brainer to some, but I could imagine some companies only reprinting for people who complain) and being transparent about what went wrong.

Leave a Reply

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