What Getting Lost Taught me About App Analytics
It’s a fact: I get lost everywhere, all the time. It doesn’t matter how often I’ve been somewhere, I can still get lost on the way. One could easily blame smartphones for robbing my entire generation of all sense of direction. I don’t know how true that is but, at least in my case, there was not much to take away in the first place. As a kid, I could easily get lost while walking to a friend’s house, just a block or two away. Back then, I cared a whole lot more about what people thought so I would never ask anyone for directions. Worse, if anyone asked if I was lost, I would pretend I was just taking a nice walk around the neighborhood to avoid any embarrassment. I would come back home, exhausted and hungry, pretending I had a good time at my friend’s.
Now, imagine that instead of a kid trying to get to a friend’s house, I’m a company trying to make a mobile app successful. Instead of a house with an address, my destination is marked by a certain download count, user retention rate, customer lifetime value or any combination of relevant indicators. The road is long and arguably hard to figure out. No matter how confident I am, if I don’t regularly make sure I’m still on the right way, I have no chance in hell to get there! Well, developing an app without proper analytics is pretty much like pretending you know your way when you really can’t be sure.
There is another bad habit that I had as a kid: I always carried a backpack with way too much stuff in it. I didn’t take the time to see if I needed this textbook or that pencil case, I just carried everything around, all the time. But apps are just the same: when they are too heavy, they slow you down. In more concrete terms, any app feature has to be maintained and tested regularly as the world around it keeps evolving. All that work is pure waste if users don’t really need that feature, so you really want to know what features bring your app closer to success and part with those that don’t. The best way to do so is with split testing: when introducing a new feature, only give it to a subset of users. If those users are measurably happier, more engaged or otherwise show that your app is closer to success, then keep it. If they don’t, try something else but don’t carry the dead weight like it’s gold. It’s not.
Now, I’m a grown up and I’d like to say that my sense of direction has improved. It hasn’t. Since I travel a lot, my friends always ask me how I manage not to be lost all the time. The answer, 95% of the time, is Google Maps. Yet, as we know, the map is not the terrain (or, more likely, my phone’s battery is dead) and sometimes I need to ask for directions. Then, I try not to let my ego lead the way.