Friday, 22 August 2014

A Case Study of What Went Wrong With my App

One of my apps, Fuel Buddy, has not been doing well lately. And by doing well, I mean the downloads have been low, and rankings are on a steady decline. To take corrective action I took some desperate measures like adding a new feature, changing screenshots, etc. But there was no improvement. So I decided to analyze thoroughly what the problem could be. I gave myself an entire day to figure out what could be going wrong. What I found was surprising to say the least. The reasons were a combination of multiple points and not just one. I took note of my findings to make sure I don't repeat the mistakes again. Sharing them here.

Fuel Buddy Downloads Case Study

  • This one is obvious, but needs to stated as the first point. Rankings and downloads go hand in hand. One affects the other. This one clearly has a spiral effect. The sooner you nip the bud the better.
  • A slight drop in ratings over two days plummeted my rankings, which led to a drop in downloads (see point 1). Case in point, my app got an average rating of 2 and 3 on the 6th and 7th of May. This led to a drop in downloads by almost 30%!! Though I have not seen the opposite being true, a spike in ratings leading to a better ranking. None the less, avoid low ratings at all costs.
  • Change in the app title can lead to big drop in installs and high uninstalls. This can only be experimented with over a long period, a few months maybe. Eg, I Changed Title to FuelBuddy-Car Service, Mileage, and saw the rate of uninstalls rise. Plausible reason can be that people expected the app to be more of a car service app, whereas it is more on the lines of a mileage logger. Changed the title back to Fuel Buddy - Car Mileage Log and the rate of uninstalls reduced.
  • If you see your uninstalls go up, the first thing to do is check your crashes, we tend to ignore this since they don't show up in the Developer Console, but it is a good idea to check them with a analytics software like Google Analytics or Flurry. Make sure to implement these in your app, because most people don't bother clicking the 'Send Report' button, they just uninstall the app.
The one thing I did to desperately get back on track was churn out random solutions in the hope that I fix my app. This was mainly due to the amount of realtime information available to us in the form of downloads, ratings, reviews, in-app purchases, etc. Don't get bogged down by daily data. The truth is that daily fluctuations can be caused by external factors. It all makes sense when you zoom out, and look at the big picture. Study the trend over a week or month. That will give you the real root cause which you can act upon.

Hope these points help you out. I'll be doing another case study of my Facebook Ad campaign on one of my apps. Stay tuned for that.

No comments:

Post a Comment