Your app’s screen real estate is precious, so you went to serious effort to streamline its UI and get rid of clutter. Then, you begrudgingly squeezed in an ad unit — it’s ugly, but you need to get paid. You know you should prompt for reviews of your app, but you bury the nag for ratings in your app’s “About” section because you don’t know where else to stick it.
Well, stop doing that. That’s what we used to do too, until a series of experiments showed us that prompting for reviews is the key to getting more downloads and happier customers — and it’s more valuable than any ad unit.
Instead of ads, we toss a coin to use some of our most valuable screen real estate to prompt for a review instead of showing an ad. It’s worth it to us to invest these precious pixels in the long-term health of our app’s reputation.
We experimented with different formats, measured the results for performance, and are sharing our results today.
First: why we do this
It’s not easy make money from the Store. Every dollar is hard-won. That means our decision to give up some of our most valuable ad inventory in favor of a review collection prompt needs some context.
We realized our apps’ Store listings need a positive rating and lots of glowing reviews to succeed. User feedback affects every part of an app’s conversion funnel, starting with its appearance in the Store’s search results and ending with the Download page.
And although it’s hard to know for sure, we have strong evidence that an app’s rating affects its ranking in Store listings for keywords and categories.
So it’s clear that to succeed in the long run, we need to get as many reviews as possible from our users. But we didn’t want to interrupt our apps’ workflows with intrusive modals, something that’s so common (and annoying) on iOS that Apple asks its developers to use a rate-limited native controller.
That makes our ad banner space the ideal location for a gentle prompt for feedback — even if it does cost us money.
How to make the ask
We’ve experimented with a dozen formats over the past year. Here are the two that work best in our apps.
The first is dead-obvious: a simple row of stars, asking the user to click one to review the app.
Note that we keep the “AD” disclosure in the prompt. That’s because we want users to know this prompt isn’t part of the app’s UI. (This disclosure links to our website with a support/complaint email.)
Here’s a sample of the performance of this format:
You can see it does very well: 2.3% of users who see this prompt will click on it.
No matter what the user clicks, they’re sent to the Store’s review experience. Does that sound risky? It’s not. As you can see, at least 75% of the users who click this prompt choose the fourth or the fifth star.
You might think it’s still worth segmenting clicks on the low stars from clicks on the high stars, and sending the low stars elsewhere. But if there’s a problem with the app, it’s better for us to hear about it than not, even if it temporarily costs us some points on our review average.
If we had shown an ad in this space instead of the ratings prompt, we could have earned some decent money. At a $0.50 CPM, this prompt “cost” us $640 to run. But spending $640 in exchange for more than 22,000 users potentially leaving positive reviews (against 7,200 who clicked through with neutral or negative intent) is well worth the lost earnings to us.
Although this format performed fine, we’ve moved to an even better concept that makes customers happier and drives even more positive reviews.
The second creative is a little less distracting:
Note the “AD” disclosure still appears. This prompt is softer and more conversational. It also either offers the user technical support, or asks for a review, depending on which face they click. Since we ask the user for consent before sending them to the Store’s review experience, the segmentation doesn’t go against Microsoft’s rules for ratings manipulation. It also gives us a chance to help users who really just need technical support.
Here are this format’s stats. A “click” is counted whenever someone clicks through on either the tech support link or the review link.
Even though the click-through rate is lower, this format only sends users with positive intent to the review experience, and it never surprises users by popping up the Store. The ~20 percent of users who click through for tech support reach a form that emails our tech support team with their question. We’re usually able to solve their problem, which means the possibility of a positive review from them in the future.
You got the review — now what?
Respond to your reviews. Respond to your feedback in the Feedback Hub. Do this graciously, regularly, and for every single user, even if you have to use machine translation to do it.
Reviews with responses look better on your Store listing, they give you a chance to rebut any unfair negative feedback, and they show users that you’re listening. And in many cases, a sympathetic response to negative feedback will cause your user to rethink their low rating — you might squeeze another star or two out of them just for showing you cared.
It’s easy to differentiate yourself from other developers in the Store right now by providing excellent customer support, but that won’t always be the case. Doing the right thing for your reviewers now will pay massive dividends down the road.