Keep or Reset Your App Store Rating on a New Version? A Decision Framework

Reset your App Store rating only after a real turnaround, when your lifetime average is anchored by an old version that no longer resembles the app *and* you can mint enough fresh ratings to refill the meter fast. Otherwise, keep it. Apple does let you reset the summary rating when you ship a new version, but the button clears your star average and your rating *count* in one press, and that count is often worth more than the average. A 4.5 built on ten thousand ratings tells a browsing installer other people trust you. A clean 4.9 built on eleven tells them nobody's checked yet.

This decision only exists on iOS. Apple's rating is a cumulative lifetime average with no public recency weighting, so a rough launch can shackle an app to a low number for years, and the reset is Apple's escape hatch. Google Play took the other path and skipped the button entirely ([you can't reset a Google Play rating](/blog/can-you-reset-your-google-play-rating), and you don't need to). Below is the framework: what the reset actually does, what it costs, when it rescues you, and the move that often beats it outright.

Yes. In App Store Connect, on your app's Ratings and Reviews page, there's a Reset Summary Rating control. Trigger it, confirm, and the reset takes effect when your *next* app version goes live, not immediately. Once that version releases, the star average and rating count shown on your product page zero out, and the app starts accumulating a fresh summary from the new build forward. Your existing written reviews aren't deleted; they stay readable in the reviews list, but they stop feeding the summary star average.

It wasn't always a choice. Before iOS 11 (2017), App Store ratings reset automatically every version; Apple then consolidated them into one cumulative average and handed developers the manual reset as the opt-in replacement. Google made the opposite call at I/O 2019, moving Play to a recency-weighted rating that heals itself as new reviews arrive, which is why Android has no reset switch and iOS does. [How your app's star rating is actually calculated](/blog/how-is-your-app-star-rating-calculated) puts both stores' math side by side.

The average is only half of what a shopper reads. The count is social proof, and it converts. Apptentive found that moving an app from three stars to four lifts conversion by roughly 89%, and a large, believable sample is a big part of what makes a high rating persuasive. Reset, and you trade a trusted 4.5-on-10,000 for a fragile 4.9-on-nothing; for the first days after release your page shows a thin count or a "not enough ratings yet" state that suppresses installs right when you were trying to look better.

So the cost isn't abstract: you're spending years of accumulated trust for a cleaner headline number, and you only come out ahead if the old average was genuinely dragging you *and* you can rebuild volume quickly. Measure twice.

There's a narrow, real case for it. Reset earns its keep when the lifetime average describes an app that no longer exists:

Most of the time, the honest answer is don't. Keep your rating if any of these is true:

Decided the turnaround is real? The reset takes effect on release, so line up the ratings *before* you pull the trigger, not after, and you're never showing an empty product page longer than you have to.

Before you spend your rating count, remember the cheaper lever: convert the unhappy users you already have. Replying makes a reviewer meaningfully more likely to revise the rating upward. A study of 4.5 million reviews by Hassan et al. found people roughly six times more likely to raise their rating after a developer replied (4.4% versus 0.7%), McIlroy et al. (IEEE, 2017) found 38.7% of post-response rating changes were increases, and Google's I/O 2019 guidance cited an average +0.7 stars for developers who respond. That works on iOS too, and every complaint you convert is a one-star that never anchors your cumulative average in the first place ([does replying actually raise your rating](/blog/does-replying-to-app-reviews-raise-your-rating) breaks down the causation).

So for most apps the sequence is: reply, fix, and re-earn *first*. Reset is the last resort for when that math genuinely can't win. Here's the kind of reply that earns a re-rate on the App Store:

That reply names the build that broke, the build that fixes it, and a path back if it doesn't. Apple publishes no official character limit for developer replies (community testing suggests a few thousand characters), so you have room, but tight and specific beats long. Writing one is easy; writing one for every reviewable review, fast, in the reviewer's own language, across a launch week when you're also shipping the fix, is where teams fall behind — and slow replies harden into permanent one-stars. That gap is what [ReplyArgus](/features) closes: it watches your App Store and Google Play reviews in one inbox, drafts a grounded on-brand reply for each (sized to each store), and lets you approve in a click or opt into rule-based auto-publish for the clean cases.

The reset button is real, but it's rarely the answer. Fix the app, reply fast, and re-earn your ratings before you spend the count you've built — and reserve the reset for the true turnaround where the old average is dead weight and you're ready to refill the meter. [Start free with ReplyArgus](/signup), no card, and Argus drafts your first reply in minutes, in the reviewer's own language and sized for each store.