The screenshot test that kept losing
Most teams assume that better design drives better conversion. That assumption costs downloads.
Super Unlimited VPN, the #1 VPN app globally by downloads, has run hundreds of screenshot A/B tests over five years. Eighty percent of modern redesigns lost to the original layouts. The new versions featured updated colors, contemporary design trends, and polished visual hierarchies. Users preferred what they already recognized.
The finding is counterintuitive but consistent: for high-traffic apps already ranking near the top, visual familiarity often beats aesthetic improvement. The risk profile is asymmetric—disrupting a proven asset is more dangerous than the marginal lift a redesign might deliver. Data overrides design judgment when the stakes are live traffic.
This lesson applies beyond VPN apps. Industry data from AppTweak shows that 57% of top games on Google Play test wiki:screenshot variations at least twice per year. Most apps on the Apple App Store test fewer than four times annually. The opportunity is in frequency and rigor, not in chasing trends. When conversion is the target, what users respond to matters more than what looks polished in a design review.
Custom Product Pages entered organic search—and everything changed
Until July 2025, wiki:custom-product-pages-cpp on iOS served one purpose: directing paid campaign traffic to tailored landing pages. Apple's keyword linking update fundamentally altered that role. CPPs now appear in organic search results when users query keywords tied to specific pages.
This shift unlocks intent matching at the organic level. A fitness app can show running-focused screenshots for "run tracker" and strength training visuals for "workout log"—different users, different queries, different pages. All organic.
Apple increased the CPP limit from 35 to 70 pages per app. The expansion creates room for audience segmentation, seasonal campaigns, and feature-specific funnels that were previously impossible without splitting traffic into separate app listings. Several mechanics remain under-tested: how Apple handles keyword overlaps between CPPs, whether query combinations work or only single tokens, and how CPPs compete with the default listing when keywords overlap.
The practical implication: teams still treating CPPs as a paid-only tool are missing a structural shift in how iOS discovery works. Organic strategy now requires page-level segmentation, not just metadata iteration.
Google Play shifted from installs to retention—and rankings followed
The algorithmic change most teams have not yet adjusted for came from Google in 2025. Install volume is no longer the primary ranking signal. Retention and engagement now carry more weight.
Apple's App Store Transparency Report showed that redownloads outpace new downloads by more than 2x—839 million new downloads per week versus 1.9 billion redownloads. Platforms see this and respond. Google introduced the You tab, Collections on the Android home screen, and the Level Up program for games that hit engagement benchmarks. Each feature rewards apps that keep users, not just those that acquire them.
In practice, this means acquisition and retention can no longer be optimized in isolation. wiki:in-app-events on iOS and promotional content on Google Play serve dual purposes: attracting new users and re-engaging those who left. If users churn quickly, the algorithm notices. Organic performance and retention are now causally linked.
The shift also affects how teams interpret A/B test results. A variant that increases installs but decreases retention rate may register as a win in platform dashboards but deliver a net loss in algorithmic favor over time. Full-funnel measurement is no longer optional.
Keyword strategy is not what it used to be
Keyword optimization remains foundational, but the mechanics have changed in ways many teams have not yet internalized.
On iOS, the character limits remain strict: title, subtitle, and the hidden keywords field. The goal is to cover as many unique terms as possible without duplicates, because Apple combines them within a locale. The description is not indexed—it works only for conversion. At WWDC 2025, Apple announced AI-generated App Store Tags, created from app metadata including screenshots. These tags affect browse placements, not search result ranking, but they introduce a new surface area for discoverability outside traditional keyword fields.
Google Play indexes the title, short description, and full description. keyword ranking depends on organic keyword density—roughly one exact match per 250 characters. Overstuffing hurts rankings. Since 2025, Google added Guided Search, which organizes results by user intent rather than literal query matching. Users increasingly type goals ("find housing") instead of keywords ("real estate app"), and the algorithm sorts apps into categories. metadata optimization now requires thinking about user intent, not just query strings.
The practical shift: long-tail queries with lower search volume but higher specificity convert better than broad category terms. "Remove background from photo" brings more qualified traffic than "photo editor" because the user already knows what they need. Teams optimizing only for high-volume head terms are leaving conversion on the table.
What the webinar revealed about ASO myths that persist
A recent AppFollow webinar featuring ASO leads from MY.GAMES, TapNation, and Toca Boca surfaced several widely held assumptions that do not survive contact with data.
The first: blaming the algorithm when downloads drop. Most drops are not caused by algorithmic changes. They come from competitor bidding shifts, category changes, or seasonal patterns. Teams that default to "the algorithm changed" skip diagnostic work—splitting data by traffic source, checking browse versus search performance, and tracking competitor movements. The algorithm is a convenient explanation because it requires no follow-up action. Real root causes are findable if teams look.
The second: ranking for more keywords equals success. A growing list of ranked terms looks good in a slide deck. It demonstrates activity more than results. If the keywords have low traffic and thin conversion, the number is just a number. Relevance determines whether a keyword is worth having. keyword strategy should track impressions alongside installs and conversion rate. When impressions go up and conversion drops, something is off.
The third: creative testing without hypotheses. Running screenshot tests without knowing what you're trying to learn produces a history of tests but no accumulated understanding. Before running a test, teams should know what they're testing, what a good result looks like, and what they'd do with a negative result. Testing whether screenshots "look nicer" is not a testable idea. Testing whether showing a new feature in the first two frames improves conversion is.
One simple tactic worth trying: convert screenshots to black and white and see where your eye lands. When there's too much competing for attention, the black-and-white version makes that obvious. Effective store creative focuses on one or two things clearly. Screenshots that try to communicate everything at once often communicate nothing particularly well.
Apple Search Ads and ASO are not separate budgets—they're one feedback loop
Most teams treat Apple Search Ads and ASO as independent channels with different owners. In practice, both work on the same page in the same store. When there's no connection between them, both lose efficiency.
Platform data from Apple shows that with the March 2026 expansion of additional ad slots in search results across all markets, the risk of cannibalization increased. Paid budget grows, paid installs grow, but total results barely move—because ads replace organic traffic instead of adding new users. Simple rule: if an app ranks organically in the top 1-3 for a query, aggressive bidding on that same keyword requires explicit justification. If the organic position is below top 10, paid coverage likely delivers real incremental lift.
The more valuable use of apple search ads: testing keywords before committing metadata. In organic, verifying one hypothesis takes 2-4 weeks and requires a metadata iteration. Through ASA, the same data arrives in days. Launch a campaign with exact match on target keywords and watch tap-through rate and conversion rate. Keywords with high conversion in paid campaigns are candidates for the title, subtitle, or keyword field. Keywords with low conversion signal that the page does not match user expectations.
Teams that use ASA data to inform metadata decisions get faster, cleaner feedback loops than those optimizing in isolation. The platforms are converging: Apple deepens personalization through CPPs and contextual search, Google builds a re-engagement ecosystem where the store is a persistent touchpoint, not just a download gate. The shared signal from both: new installs matter, but user retention matters more.
The foldable device shift and what it means for app optimization
Apple's entry into the foldable market in 2026 is projected to capture 46% of North American foldable market share, displacing Samsung (from 51% to 29%), Motorola (from 44% to 23%), and Google Pixel (from 5% to 3%). The structural reallocation follows a predictable pattern: upgrade demand within the existing iPhone user base plus replacement demand from Android foldable users.
For ASO practitioners, this matters because device form factors influence how users evaluate apps. Foldables expand screen real estate, change how app preview video and wiki:screenshot content displays, and shift user expectations around app functionality. Apps optimized for traditional phone screens may underperform on foldables if visual assets do not adapt to larger, unfolded displays. Testing store creative on foldable simulators is not yet common practice. It should be.
What actually works in 2026
The gap between adequate ASO and effective ASO is wider than most teams realize. Keywords are in, screenshots exist, the app is live, and someone can point to a ranked keyword chart. That's table stakes. What separates performance:
- Testing frequency over design instinct. Data beats opinions. Five years of A/B testing at scale shows that what users respond to often contradicts what looks good in internal reviews.
- Page-level segmentation. CPPs are no longer optional for iOS apps with meaningful traffic. Intent matching at the organic level is now possible—teams not using it are leaving conversions on the table.
- Retention as a ranking input. Google made the shift explicit in 2025. Apple signals it through redownload metrics. Acquisition and retention are causally linked in algorithmic ranking.
- ASA as a testing ground, not just a budget line. Keyword performance data from paid campaigns informs metadata decisions faster and more cleanly than waiting for organic feedback loops.
- Localization beyond translation. Translating text while leaving screenshots in English loses conversion in markets with low English proficiency. Visual assets need localization advanced moc as much as metadata does.