6
3 Comments

Controlling Android app outcomes with ratings

Our app relies heavily on the Bluetooth protocol. When the app was designed and developed, I wanted a large potential target market. So I expected the developers to help us support as many Android versions as possible.

What happened is that even early on, we started getting 1 star ratings and reviews like, "not working" or "doesn't work" but we were also getting lots of 5 star reviews of people saying "awesome app".

I thought it was bad code but I finally figured out that it's about the relationship between the Bluetooth chipsets built into certain hardware platforms and the Bluetooth protocol. That is to say, certain chipsets on older devices don't support specific versions of Bluetooth (Classic if you know much about Bluetooth).

So, what to do? My only real option to stop the bleeding is to look at cumulative ratings by device and use the Google Play Console to exclude poorly performing devices on the premise that they aren't a good fit for our app. If someone is using one of those devices, they can't install our app. I think this is the correct action because really, the problem is not our app.

Even after excluding 58 hardware devices totaling 2204 installs with a cumulative average rating of 2.07, we still support 12231 hardware devices which are responsible for 22368 installs with a cumulative average rating of 4.04.

  1. 1

    That's the right way to go about it. Due to the hardware differences in various Android devices, it's impossible to test it for all the devices and some manufacturers are notorious about certain issues and restrictions.

    It's not just related to ratings but also about functionality. There's no point in keeping the app available if the device doesn't support it.

    I had the same issue with one of my open source apps. The app's main functionality used accessibility service and worked great. But then one day I started getting complaints and poor reviews that the accessibility service was getting stopped after every launch.

    It turned out to be some restriction on MiUi and Xiaomi devices. There was nothing I could do to fix that. It was a system restriction. I had to de-list all the Xiaomi devices.

    1. 1

      Honestly, I was a bit nervous about it. I was initially concerned it would cost us users but then I convinced myself that it was really the only solution - and the right one.

      By the way, do you know what kind of trial conversion rate I might be able to expect? I've read some really conflicting numbers in trying to research trial to paid conversions.

      1. 1

        Paid conversions depend on the app category, value it provides, pricing and many other factors. I don't think there's one answer that fits all.

Trending on Indie Hackers
How I grew a side project to 100k Unique Visitors in 7 days with 0 audience 47 comments Competing with Product Hunt: a month later 33 comments Why do you hate marketing? 27 comments $15k revenues in <4 months as a solopreneur 14 comments Use Your Product 13 comments How I Launched FrontendEase 13 comments