The app store optimization playbook
keywords, conversion assets and the Apple featuring pipeline for consumer AI apps.
last updated 30 August 2026
app store optimization is the compounding free channel for consumer apps. while paid CPIs climb every quarter, a listing that ranks for the right searches and converts the people who land on it delivers installs every day at zero marginal cost. the work splits into three jobs founders usually blur into one: ranking, converting and getting featured. run them separately, on a fixed cadence, and the channel compounds.
Three jobs, not one
ASO is three different jobs, and most founders fail at it because they treat it as one. ranking is a keyword problem: which searches you show up for. converting is an asset problem: whether the screenshots, icon and ratings turn a viewer into an install. featuring is an editorial problem: whether Apple’s or Google’s human editors decide your app deserves a spotlight. each job has different levers and different feedback loops. keyword changes show movement in days. asset tests take weeks to read. featuring runs on editorial calendars planned months out. when you mix them, you change five things at once and learn nothing from any of them.
Why AI keywords are brutal real estate
the head terms every consumer AI app wants are already owned. searches like “ai photo editor” or “ai chat” are held by apps with years of download velocity and review counts you cannot match in your first year, and the store rewards exactly those two things. the honest entry point is the long tail: specific phrases with lower volume, higher intent and thin competition. rank first for twenty specific searches, then let that history earn you a shot at the generic ones. it is the same logic as bidding in paid ads: you do not outbid the biggest budget on the broadest term, you find the pocket it ignores.
Converting is the half founders skip
ranking gets you seen, the listing gets you installed, and the second half is where most of the easy gains sit. a viewer decides in seconds, off the first two screenshots and the star rating, whether your app is worth the tap. improving that decision lifts every channel at once, which is the quiet superpower of this work. i ran growth for ZuAI, a consumer AI study app that scaled 10K to 2M users at a $0.02 blended CAC, and every channel in that mix ended at the same store listing. its conversion rate multiplied everything upstream of it, paid and organic alike.
Featuring is a pipeline, not a lottery
Apple and Google both run real editorial processes, and you can apply to them. featuring is not a growth strategy on its own, the spike is short, but it is free velocity, credibility you can point at, and a deadline that forces the rest of your store presence into shape. treat it as a recurring pipeline tied to your release calendar, not a prayer.
The cadence i run
monthly keyword review, quarterly asset test. once a month, check what you rank for, what moved, and which new phrases show up in reviews and autocomplete, then adjust the metadata. once a quarter, run one clean asset experiment: a new first screenshot, a different message order, a preview video. and every major update, submit a featuring nomination. that is the whole system. it fits in a few hours a month, which is exactly why it survives contact with everything else a founder has to do.
The honest tradeoffs
ASO will not save an app that needs users this week. it compounds over months, it cannot fix weak retention, and it only pays back if people are actually searching for what you built. if your launch window is now, run the launch playbook first and let the store work catch up behind it. but if you plan to still exist in a year, this is the cheapest traffic you will ever buy, because you buy it once with work instead of forever with budget.
the paid side that runs alongside this lives in the paid ads playbook. the breakdown call is free. Let’s talk Growth.
In this playbook
How to get featured by Apple
Apple runs a real nomination form. how to submit it 6 to 8 weeks ahead, what editors actually pick, and what a feature realistically does.
App store optimizationASO keyword research when every AI keyword is a war
why head terms are unwinnable for new AI apps and the long-tail method that works: mine reviews, autocomplete and competitor names for intent.
new chapters land here as they are written. the breakdown call gets them first.