Discover more from Dan’s Media & AI Sandwich
App Store ratings are broken
(and some thoughts on how to fix them)
App Store ratings matter. According to Apptentive, 90% of users consider star ratings essential in deciding whether to download an app and 50% say they will download a 3-star app vs 95% for a 4-star app.
What a difference a star makes
Ratings also affect your app’s ranking in App Store results, with apps averaging above 4 stars being upweighted and those averaging below 3 being downweighted1.
Unfortunately, your app won’t naturally receive ratings from a representative cross-section of users, as we are predisposed to be more motivated to share negative experiences2.
Early app developers quickly realised this and started introducing rating prompts into their app’s user experience to try and encourage the silent (and most likely contented) majority to leave a review.
Apple imposed some standardisation with iOS 10.3, requiring developers to use the official in-app rating UI. However, exactly when and how you trigger the official rating UI has become an art form all of its own.
For example, by analysing user behaviour you can introduce logic to only show the ratings prompt to your most engaged users at a moment in their user journey when they are likely to be most predisposed to giving your app a favourable rating.
In the same way as being good (or bad) at job interviews doesn’t necessarily mean you’ll be good (or bad) at the job you’re interviewing for, being good (or bad) at when and how to prompt users to review your app can potentially be a bigger determinant of your app store rating than the quality of the app itself.
Another weakness of the App Store rating system is that it’s pretty much the only way your users can express their frustration at a missing feature or bug and be heard. Like a newborn baby who only has one way of getting attention (crying), I’ve left one star ratings (with an accompanying constructive review) of apps which I generally rate highly because an element of the experience is broken and I want to get the developers’ attention.
I was on the sharp end of this equation in 2018 when we introduced BBC Sounds as a successor to BBC iPlayer Radio. Many devoted iPlayer Radio users were unhappy with the change and expressed their frustration with a 1 star review.
It was clear from reading the accompanying reviews that not all of those users had actually used the BBC Sounds app, whilst others had and were incandescent about the absence of a single feature that would be introduced before they were migrated over from iPlayer Radio.
Another issue which blights pretty much all manifestations of user ratings where there’s money at stake (with Amazon the poster child) is app ratings for sale. I won’t give them the Googlejuice by linking to them, but check out the below screengrab from the top organic search result for ‘app store reviews for sale’.
So, lots of problems, Dan. What about some possible solutions?
Below are a few ideas. I’d be interested to hear other people’s, especially developers and ASO specialists.
1.) Only let people who have used the app rate it
Simple to implement, this would make it harder (though not impossible) for those selling paid-for ratings (as those posting the bogus ratings would need to actually use the apps before they could leave a review) and reduce the number of trigger happy ratings from regular users.
Spotify have done a nice job of implementing this for podcasts (see below grab).
But what about apps which crash on opening or which try to scam you? Shouldn’t you be allowed to rate those without using them? I’d suggest that’s what the Report a Problem/Flag as Inappropriate links are there for, putting the onus on Apple/Google to take action.
2.) Provide an alternative mechanism for users to report bugs / make feature requests
This would require more product development and user testing, but a mechanism for engaged users to be able to report bugs and make feature requests without resorting to a 1 star review or an email into the abyss would benefit users and developers alike.
A way of up and down voting bugs and feature requests would make it easier for developers and users to see what issues engaged users care most about, without the need for thousands of duplicate reviews raising the same point.
3.) Encourage users to leave reviews of the apps they use the most when they’re not trying to complete a task
One of the issues with the current implementation of App Store rating prompts is that they usually get in the way of a task you are trying to complete, leading most of us to instinctively hit the ‘Not Now’ button.
Why not instead invite users to review the apps they use the most when they’re more likely to have time to do so (e.g. via an OS notification or, dare I say it, email)?
Whilst there’s clearly no silver bullet when it comes to addressing the deficiencies of App Store ratings, it would be good to see Apple and Google prioritise improvements in this area in view of their impact on apps of all sizes.