Keywords get you seen, but creatives decide the install. In the first episode of ASO Creative Teardown, Applica’s Mladen Crnomarković scans the first screenshots of the top 15 EdTech apps, then tears down Duolingo, Mimo, and Lumosity: what a user understands in the first 3 seconds, where the message leaks, and what we’d change and why.
Performance Creative Strategist
Mladen Crnomarković is a Performance Creative Specialist at Applica, focused on ASO and UA creatives for mobile apps and games. He designs icons, screenshots and video ads, and A/B tests them on the App Store and Google Play to turn storefront visits into installs.
LinkedInMost ASO conversations start with keywords. But keywords only get you seen — creatives decide the install. Even if your app ranks first, a competitor in third place with a clearer, more eye-pleasing screenshot set can take the install from you.
That's the idea behind ASO Creative Teardown, Applica's new monthly live series. In each episode, Mladen Crnomarković, Performance Creative Specialist at Applica, takes one app category, scans the first screenshots of its top 15 apps, and tears down three of them in detail. Creatives only: icons, screenshots, and preview videos — no keywords, no metadata. Episode 1 covered EdTech.
Every app gets the same four beats:
Message hierarchy
Whether the core promise lands across the first three screenshots, the most convertible ones
The 3-second test
What a stranger understands from the icon and the first three screenshots as the app shows up in search
Versus
What the category does that this app is missing
The fix
What we'd change and why
Each app ends with one "steal this" line you can apply to your own listing.
Across the top 15 EdTech apps, the patterns were clear:
That last number is the biggest gap in the category. At Applica, the outcome is the most important thing in a screenshot set: what can people actually do or learn with your app?
Even the category leader has room to improve. Duolingo skips mockups and headline blocks, shows clean, full-bleed app UI, and places captions at the bottom styled like its own buttons — the only app of the 15 doing this. That keeps the set on-brand and makes it read like the app, not an ad.
What we'd change: Duo, one of the strongest mascots on the App Store, only appears on the sixth screenshot. The strongest outcomes — practicing conversations, using math with confidence — sit too far down the set, while chess takes the third slot. Our fix: bring Duo and the outcomes into the first three screenshots, keep the default page language-first, and move chess and math to Custom Product Pages. Interestingly, Duolingo's Google Play listing already does much of this.
Steal this: make your captions look like your app, not like an ad.
Mimo has a visually strong set, but the message gets confusing. The title says Python, while the second screenshot lists ten languages and tools at once. One phone is split across two frames, which splits the message too. And the robot icon is friendly, but it doesn't say "code" — Mimo isn't Duolingo, so the mascot only works for people who already know it.
Mimo's strongest message is the career: "kickstart your career in just 5 minutes a day." We'd lead with it, show what users actually build, keep Python and AI on the default page and move other languages to CPPs, give every frame a headline, and test an icon that signals programming. On Google Play, Mimo already leads with the 5-minutes-a-day line and adds more social proof — worth bringing to iOS.
Steal this: put the outcome and the effort in one line.
Lumosity is the one app of the three that leads with a video — but it opens on the logo instead of a hook. Its serif headlines are beautiful but hard to scan, and people on the App Store scan rather than read. After the second screenshot, the same frame repeats with only the headline changing, so nothing stops the swipe. Progress tracking only appears on the fourth screenshot, after most users have stopped scrolling.
Our fixes: a readable headline font, the phone pulled up, a broken template, numbers like "40+ games" as static proof, progress moved into the first three, and a video that opens on an outcome or a question.
Steal this: give progress a number per skill.
Strong badges like Editor's Choice, App of the Day, and millions of ratings work — but newer apps don't have them. Mladen's alternatives: your overall rating, or concrete content numbers like lessons, levels, or activities. One caution from Lumosity: a big user number next to a small ratings count can look off to users.
Most EdTech apps tour their features well, but almost nobody tells users what they'll be able to do. Put the outcome in your first three screenshots. Next episode: Health & Fitness — drop your app link in the comments or reach out to Mladen on LinkedIn to get featured.
Thank you for subscription
to Applica Digest!
The latest mobile growth insights will be
delivered straight to your inbox
