How we solved our lack of app reviews
App Store ratings matter. A lot.
App store ratings and reviews are the lifeblood of any mobile app. They are your app customers’ first impression, their most trusted barometer of quality, and their most powerful means of expressing either happiness or frustration.
- 70% of people read at least one review before downloading an app
- 75% identify reviews as a key driver for downloading an app
- 42% view app store reviews as equally or more trustworthy than personal recommendations
- 13% of people read at least seven reviews when considering a free app
The Student Beans apps had been live for 9 months, with a substantial amount of downloads, but a dismal amount of reviews. It’s human nature not to leave reviews unless you have had a negative experience. It’s a way to vent frustration.
When Apptentive broke down their data, they found that Unprompted reviews are 56% more likely to be based on negative experiences than positive ones.
Please note that this work was carried out before the native review prompt in iOS — learnings still stand for android 👀
Our solution was simple; Prompt users to leave a review!
- Where would a review prompt be best placed within the app?
- When would be the right time? (consider frustration of a prompt, could it interfere with a user goal?)
- Could negative feedback be caught before it got to the app store?
- How would we best communicate with the user to get them to review? Language is important.
Just asking people to rate your app is ~5 to 10x less effective than starting a conversation about whether or not the consumer is happy.
I wanted to create two paths for our reviewing users;
- A Positive path — which would take them to the the app store, and
- A Negative path — which would take them to our own feedback form.
This would help us to prevent negative feedback from hitting the app stores and provide proper support for their issues.
To do this we had to start the review prompt with a conversation.
👀 Potential Direction
When would be the right moment to show a review prompt?
I considered that the ultimate moment of happiness for our user would be at the point that they successfully redeemed a student discount. Playing on this success, we would have our ratings flooding in right?
However, after further consideration into the user flows we concluded that this wasn’t the right choice. Or at least not the right place to start with. It was too easy for the prompt to interfere with the main objective of our app.
Direction 1 = 🙅🏻
Having thought about how disruptive a prompt could potentially be I settled on a different direction. I decided that the first iteration should have a tiny footprint. This would allow us to dip our toes in the water, play it safe, and let us look into the data afterwards to decide on the next iteration.
Now that the prompt isn’t extremely disruptive we are able to be a little more relaxed on the logic — the triggers for when it should appear. We looked at usage rate and successful funnels to help determine when it should appear.
I used Principle to create a prototype of the feature. Principle is really simple to use and interactions can be created in little to no time at all. We could mirror this to our real devices to see exactly how it would behave, tweaking it instantly. This gave us a sense of how it would feel for the prompt to appear while using the app, as well as being able to show the developers exactly how it should work.
🙏🏻 The Results
It has been 3 months since the review prompt went live on both apps and we have seen the following positive results:
The conversational approach also encouraged those that weren’t particularly enjoying the app to actually voice their opinion. Those that weren’t annoyed enough to leave a negative review, but definitely had some gripes to get off their chest. This feature is now contributing to the prioritisation of our backlog of enhancements.
These positive results are not the end of this small feature. We use Mixpanel to give us a more in-depth look into how our users interacted with the prompt. We will use this information to improve on the current logic. Tweaks will be made based on this and effects will be reviewed on a rolling cycle until we believe we have done all we can.
Thanks for reading! 🙏
Feel free to tweet at me with any thoughts or feedback you have, or comment below. Would love to hear from you!
There will be lots more stories coming in the near future from each area of the Engineering Team. So make sure to follow for updates :)