What a good app store conversion rate actually is
Benchmarks are the wrong thing to chase. Here is what to compare your store conversion against instead, and the one change that moved installs 23% on the same spend.
the short answer: there is no single good number. a 30% conversion rate from branded search and a 30% conversion rate from a cold TikTok ad are not the same result, and averaging them together hides the only thing worth knowing. compare each traffic source against itself over time, not against an industry median.
Why the benchmark question misleads
app store conversion is the share of people who view your product page and install. the problem is that the people arriving are not one audience.
someone who searched your app by name is most of the way to installing before the page loads. someone who tapped a cold ad has never heard of you. someone browsing a category is somewhere in between. these convert at wildly different rates on the same page, so a blended number moves whenever your traffic mix moves, even when the page itself has not changed at all.
i have watched an account’s “conversion rate drop” that was entirely a mix shift: paid scaled up, branded search stayed flat, blended conversion fell, and the page was performing exactly as well as the week before. nothing needed fixing. a week was spent looking for a problem that did not exist.
What to compare against instead
three comparisons, in order of usefulness.
the same source last month. this is the only honest like for like. branded search against branded search, one ad campaign against itself. if that number moves, something real changed.
your own best source. branded search is roughly your ceiling, because it is the warmest traffic you will ever get. if cold paid converts at a fifth of branded, the gap is your page failing to do the job your brand does.
the step before, not just the install. impressions to page view to install are separate problems. a low install rate with healthy page views is a page problem. a high install rate with almost no page views is a discovery problem, and rewriting screenshots will not fix it.
The change that actually moved the number
custom product pages, which let you serve a different set of screenshots and copy per campaign instead of one page for everyone.
running them against a single generic page produced 23% more installs on the same spend. no extra budget, no new creative shot for it, just the page matching the promise the ad had already made. the full breakdown is in custom product pages.
that result is the whole argument against chasing a benchmark. the generic page was converting at a perfectly respectable rate. it was respectable and leaving a fifth of the installs on the floor at the same time.
The order I would work in
screenshots first, specifically the first two, because most people never scroll. then the icon, which is the only thing carried into every browse surface. then title and subtitle, which do double duty for keywords and for the three seconds someone spends deciding.
reviews matter too, and the only durable way to get them is to ask people who already came back, not to buy them. the shortcuts get apps removed.
if you want to put your own numbers in and see what a change is worth before you make it, the app store conversion calculator does the arithmetic. the rest of the method is in the app store optimization playbook.
want the honest read on your own store numbers? the breakdown call is free. Let’s talk Growth.