According to Apple, 70% of App Store visitors used search to discover apps in 2022. Almost 65% of downloads happened right after a search (Apple Ads, App Store, 2022 data). Your listing is judged in a results list before anyone reads it. Yet most "ASO checklists" are topics you can't fail. "Optimize your title" isn't a check.
Teams audit by feel, then re-audit with a different list, so nothing is comparable. This ASO audit checklist has 60 checks across both stores, each with a pass threshold and a method, weighted into a repeatable score. New to ASO? Start with the free-tools walkthrough; the audit then feeds the six-step ASO strategy loop.
An audit works only when every check has a pass threshold. These 60 checks score into a 100-point rubric, and 4 gates cap your grade. One is a rating under 4.0: around 95% of featured App Store apps sit at 4.0 or higher (AppTweak, 2026). 57 of the 60 checks need no paid tool.
Key Takeaways
- 60 checks, 8 weighted categories, 100 points, each scored 0, 1 or 2; 57 need no paid tool.
- Four gates cap the grade at C: brand not #1, a Play policy violation, a rating under 4.0, and Android vitals over threshold.
- 8 checks are new for 2026, including iOS 27 creative assets, CPP keywords and Ask Play.
What is an ASO audit, and what makes a checklist usable?
An ASO audit is a structured review of your whole store presence: keywords, text metadata, creative, ratings, localization and the newer discovery surfaces. Each part is scored against a threshold, so the result is a ranked fix list, not a pile of opinions. The audit is a snapshot; a strategy is the plan.
Audits have to repeat because listings move. In 2025, 34% of the top 1,000 U.S. apps on both stores updated their description twice or more (AppTweak, Trends & Benchmarks Report 2026). January's audit describes a store that has changed by summer.
Every check below has three parts: what you're checking, the pass threshold, and how to check it. A check without a threshold is an opinion. Some thresholds are official limits, like Apple's 30-character app name. Others are our judgment, tagged House. In the tables, Gate caps your grade, and 2026 marks a surface or metric new since October 2025.
Run the categories in order: a conversion fix on a page nobody sees won't show up, so diagnose visibility first.
How does the ASO audit scoring work?
Each of the 60 checks scores 0 (fail), 1 (partial) or 2 (pass). The checks sit in 8 categories whose weights add up to 100, and four gate checks cap the grade at C when any one fails. It's our own framework: ASO Agency's 2026 audit rubric.
| # | Category | Checks | Weight |
|---|---|---|---|
| 1 | Foundations & measurement | 6 | 5 |
| 2 | Keywords & rankings | 8 | 15 |
| 3 | App Store text metadata | 9 | 15 |
| 4 | Google Play listing | 7 | 10 |
| 5 | Creative & conversion (both stores) | 10 | 20 |
| 6 | Ratings, reviews & app quality | 7 | 15 |
| 7 | Localization & market coverage | 6 | 10 |
| 8 | 2026 discovery surfaces | 7 | 10 |
| Total | 60 | 100 |
Category score = points earned ÷ (2 × applicable checks) × category weight. Your total is the sum of the eight. A partial (1) means the threshold holds in some markets but not all, or is missed narrowly. Ship on one store only? Mark the other store's checks N/A and rescale. An iOS-only app drops the Google Play category, so its total is divided by 90 and multiplied by 100.
- A, 85–100: optimize at the margin.
- B, 70–84: clear gaps, quick wins available.
- C, 50–69: structural issues.
- D, below 50: rebuild the listing.
The four gates are B2 (brand not #1), D4 (Play policy violation), F1 (rating under 4.0 in a top market) and F6 (Android vitals over threshold). Why cap instead of subtract? A sub-4.0 rating or a suspended listing can erase gains everywhere else, and a subtraction would hide that behind a respectable number.
How the 100 points are weighted
Keywords and App Store text (15 each) plus the Play listing (10) carry discoverability, since search is the main discovery channel on iOS. Ratings (15) gate both stores.
How each check is run
We tagged every check by method: 20 run on our free tools, 25 in App Store Connect, Play Console or Apple Ads, and 12 by manual review. Only 3 need paid keyword data (B1, B6, G2), so 57 of the 60 need no paid tool.
How each of the 60 checks is run
A worked example
Here's the arithmetic on an illustrative two-store app. The numbers are invented to show the math; they aren't client data.
| Category (weight) | Points earned | Category score |
|---|---|---|
| 1. Foundations (5) | 8 of 12 | 3.3 |
| 2. Keywords (15) | 13 of 16 | 12.2 |
| 3. App Store text (15) | 13 of 18 | 10.8 |
| 4. Google Play (10) | 11 of 14 | 7.9 |
| 5. Creative (20) | 15 of 20 | 15.0 |
| 6. Ratings (15) | 11 of 14, F1 failed | 11.8 |
| 7. Localization (10) | 8 of 12 | 6.7 |
| 8. 2026 surfaces (10) | 7 of 14 | 5.0 |
| Total | 72.7: grade C (B by points, capped by the F1 gate) |
1. Foundations and measurement: can you trust your baseline?
You can trust your baseline only if it separates organic from paid traffic, and before from after. Otherwise every later check is guesswork, so audit the data before the listing. Google Play also changed its store listing metrics in mid-2026. They now count unique users who click Install, Open or Pre-register, not completed acquisitions (Google Play Console Help, 2026).
| ID | Check | Pass threshold | How to check |
|---|---|---|---|
| A1 | Baseline exported | 90 days of impressions, page views, installs and conversion rate per store, by source and top-5 market, saved before any change House | App Analytics; Play Console |
| A2 | Organic vs paid separated | Organic conversion rate calculated without Apple Ads or Google Ads traffic | Source filters in both consoles |
| A3 | Change log | Every metadata and creative change dated per store and locale, 12 months back House | Internal doc; version history |
| A4 2026 | Play metric break respected | No trend line spans the switch to unique clicks; reach metric added to the dashboard | Play Console store listing reports vs Grow users overview |
| A5 | Competitor set defined | 5–10 named competitors per top market, reviewed in the last 6 months House | Category charts; keyword results |
| A6 | Owner and cadence | One named owner; monthly visibility and conversion review House | Manual |
A4 is the trap: a trend line across the switch compares two different metrics. Google's help page dates the change to both June and July 2026, so mark the break and restart comparisons after it. The new reach metric was announced at I/O in May 2026 (Android Developers Blog, 2026).
2. Keywords and rankings: are you visible for the right terms?
Visibility comes first: if you don't rank, conversion work has nothing to convert. Most App Store downloads start with a search, almost 65% of them in Apple's 2022 data (Apple Ads, App Store, 2022 data). Brand #1 is a gate: when your own name doesn't return your app first, nothing downstream matters. If it doesn't return your app at all, work out why an app isn't showing in App Store search.
| ID | Check | Pass threshold | How to check |
|---|---|---|---|
| B1 | Keyword universe | Prioritized list (relevance × popularity × difficulty) per top market, refreshed within 6 months House | Apple Ads search popularity; paid ASO platform |
| B2 Gate | Brand keyword #1 | #1 for your own brand name in every live storefront, both stores | Keyword Rank Checker (iOS); manual Play search |
| B3 | Core terms in top 10 | At least 3 of your 10 priority non-brand keywords in the top 10, primary market House | Keyword Rank Checker |
| B4 | Top-50 coverage | At least 50% of priority keywords in the top 50 House | Keyword Rank Checker |
| B5 | Indexing verified | Every term in name, subtitle and keyword field returns the app in the top 200 within about 14 days of release House | Keyword Rank Checker |
| B6 | Winnability | No more than 30% of targets are head terms where the top-10 apps far out-rate you House | Top-10 ratings counts; paid difficulty data |
| B7 | Paid-to-organic loop | Apple Ads search terms reviewed quarterly; converting terms missing from metadata queued House | Apple Ads search terms report |
| B8 | Review language mined | Top 10 recurring feature phrases in recent reviews checked against metadata | App Review Keyword Analyzer |
The research method behind B1 lives in App Store keyword research, step by step. B5 is cheap to run: a term missing from the top 200 two weeks after release is either unindexed or irrelevant.
3. App Store text metadata: is every character working?
Every character is working only when nothing repeats and nothing breaks Apple's rules. As of 2026, Apple gives you 30 characters of app name, 30 of subtitle and 100 characters of keywords (App Store Connect Help, App information; Apple Developer, Product page). Apple's App Store Connect Help still says 100 bytes; saved-string tests show characters are what's enforced. Each keyword must be longer than two characters, can't repeat your app or company name, and can't name other apps or companies.
| ID | Check | Pass threshold | How to check |
|---|---|---|---|
| C1 | App name | ≤30 characters; brand plus your #1 non-brand keyword; not truncated in search | Metadata Character Counter; Store Page Preview |
| C2 | Subtitle | ≤30 characters; no word repeated from the name; reads as a benefit | Metadata Character Counter |
| C3 | Keyword field fill | At least 95 of 100 characters used House; commas, no spaces | Keyword Field Optimizer |
| C4 | No duplication | No word repeated from name or subtitle; no app or company name; no singular/plural pairs House | Keyword Field Optimizer; Keyword Density Checker |
| C5 | No wasted or risky terms | No other apps' or companies' names; every keyword over 2 characters; no filler like "app" or "free" House | Keyword Field Optimizer; manual |
| C6 | Promotional text | All 170 characters used in every live locale; refreshed within 90 days House | App Store Connect |
| C7 | Description above the fold | Lines before "more" say what the app does, for whom, plus one proof point House | Store Page Preview |
| C8 | Category | Primary category matches your priority keywords and competitor set; reviewed yearly House | App Store Connect; category charts |
| C9 | Promoted in-app purchases | If you sell subscriptions or non-consumables: top ones promoted (up to 20), display names ≤30 characters, 1024×1024 images | App Store Connect |
Two details catch experienced teams. The keyword field is version-level metadata, so C3 and C4 fixes ship with your next app version, while promotional text (C6) can change any time. And for C9, Apple says consumable in-app purchases won't appear in search results (Apple Developer, Promoting in-app purchases, 2026).
4. Google Play listing: is it indexed and policy-safe?
A Play listing passes when its text covers your keywords without breaking policy. Google says the text in your store listing "helps users discover your app" (Google Play Console Help, 2026). But repetitive or irrelevant keywords "can create an unpleasant user experience and result in an app being suspended on Google Play" (Google Play Console Help, 2026). A violation is a gate.
| ID | Check | Pass threshold | How to check |
|---|---|---|---|
| D1 | Title | ≤30 characters; brand plus primary keyword; no emoji, ALL CAPS, "Free", "#1" or "Best" | Metadata Character Counter |
| D2 | Short description | ≤80 characters; benefit plus a priority keyword; no price or ranking claims | Metadata Character Counter |
| D3 | Full description density | Exact target phrases 3–5 times, naturally, within 4,000 characters; any single keyword around 2–3% of the description's words House | Keyword Density Checker |
| D4 Gate | Policy-clean metadata | No keyword lists; no ranking, award or price claims in text or graphics; no "Download now" calls to action in screenshots or video | Manual review against Play policy (how to fix a failing Play listing check) |
| D5 | Feature graphic and video | 1024×500 feature graphic with no claims text; promo video public or unlisted, ads off | Play Console |
| D6 | Screenshots | At least 4 of 8 phone slots used House; first frames benefit-led; tablet shots if you support tablets | Store Page Preview |
| D7 | Category and tags | Correct category; up to 5 tags, each obviously relevant | Play Console |
Some checklists still prescribe a fixed keyword density for Play, but Google publishes no number. Our rule of thumb: keep any single keyword around 2–3% of the full description's words, repeat exact target phrases 3–5 times, and write for readers first (the Keyword Density Checker flags any word above 5%). Google's own violation example is a comma-separated list: "Car racing, car driving, race cars…" (Google Play Console Help, metadata guidelines, 2026). For D4, graphics can't say "Best", "#1", "Top", "New", "Free", "Discount", "Sale" or "Million Downloads" (Google Play Console Help, preview assets, 2026). For ranking context, see how Google Play ranks apps.
5. Creative and conversion: does the page win the tap?
The page wins the tap only when its creative works at two sizes: the search result and the product page. On iOS, up to 3 screenshots can appear in search (Apple Developer, Asset best practices, 2026). Creative is also the least-tested area. In 2025, around 90% of top App Store apps never A/B tested their icon or a video (AppTweak, Trends & Benchmarks Report 2026).
| ID | Check | Pass threshold | How to check |
|---|---|---|---|
| E1 | Icon | Recognizable at search-result size; distinct from the top 3 results for your #1 keyword | Store Page Preview; Keyword Rank Checker |
| E2 | Search-visible frames | First 3 portrait screenshots each carry one core benefit, readable at thumbnail size | Store Page Preview |
| E3 | Captions | ≤7 words House; benefit-led; written for people, not as a keyword field | Manual |
| E4 | Slot usage | At least 6 of 10 iOS screenshots House, each the next step of the story | App Store Connect |
| E5 | App preview | If used: 15–30 seconds; core benefit in the first 3 seconds, sound off House; poster frame chosen (default is 5 s) | App Store Connect |
| E6 | Conversion vs peers | Conversion rate at or above your peer median, primary market House | App Analytics peer group benchmarks |
| E7 | Search-to-page leak | Top-5 keywords tracked from impressions to installs; high-impression, low-install terms flagged | App Analytics; Play Console |
| E8 | Testing cadence | At least 2 completed tests per store in 12 months, one of them icon or video House | Product Page Optimization; store listing experiments |
| E9 | Device coverage | Native assets for every supported device class; no stretched phone shots | App Store Connect; Play Console |
| E10 | Freshness | Screenshots match the current UI; creative reviewed within 12 months House | Manual |
The testing gap: top apps in 2025
E6 uses Apple's own yardstick. App Analytics peer group benchmarks place your conversion rate against the 25th, 50th and 75th percentiles for your category and business model (App Store Connect Help, 2026). We don't quote a universal benchmark, because the ones that circulate come from gated datasets. Before you spend a test slot on E8, read what to test first, and for how long.
6. Ratings, reviews and app quality: is anything capping conversion?
Two things can: a rating under 4.0 and, on Google Play, poor Android vitals. Both are gates. Around 95% of featured App Store apps are rated 4.0 or higher, and 65% sit at 4.6 or above (AppTweak, 2026). If an app exceeds an Android vitals bad-behavior threshold, "Play may reduce the visibility of your title" and may show a warning on the listing (Android Developers, Android vitals, 2026).
| ID | Check | Pass threshold | How to check |
|---|---|---|---|
| F1 Gate | Rating floor | At least 4.0 in every top-5 market on both stores; target 4.5+ House | App Store Connect; Play Console |
| F2 | Rating trend | 90-day average at or above lifetime average House | App Review Keyword Analyzer; consoles |
| F3 | Prompt design | Native prompt fires after a success moment, never at launch or after an error; iOS allows it 3 times in 365 days | Manual QA |
| F4 | Response coverage | 80%+ of 1–2 star reviews from the last 30 days answered within 7 days House | App Store Connect; Play Console |
| F5 | Complaint themes mapped | Top 3 complaint themes from the last 90 days identified and ticketed to product | App Review Keyword Analyzer |
| F6 Gate 2026 | Android vitals | Below every bad-behavior threshold: crashes 1.09% overall, 8% per phone; ANRs 0.47%, 8%; excessive wake locks 5% (28 days) | Play Console, Android vitals |
| F7 | Review integrity | No incentivized reviews; native prompts only; no review gating | Manual |
F6 gained a line in 2026: from March 1, titles over the excessive wake-lock threshold may be excluded "from prominent discovery surfaces such as recommendations" (Android Developers Blog, 2025).
For F7: in 2025, Apple blocked close to 195 million fraudulent ratings and reviews out of over 1.3 billion processed (Apple Newsroom, May 2026). If F1 fails, start with the 4.0 rating line and how to recover from it.
7. Localization and market coverage: where are you invisible?
You're invisible wherever a market sends impressions or revenue but sees metadata that wasn't researched for it. So the audit question isn't "are we localized?" It's whether each of those markets gets native metadata built from local keyword research. The list just grew: on March 31, 2026, Apple added 11 languages to App Store Connect, bringing supported localizations to 50 (Apple Developer News, 2026). They are Bangla, Gujarati, Kannada, Malayalam, Marathi, Odia, Punjabi, Slovenian, Tamil, Telugu and Urdu.
| ID | Check | Pass threshold | How to check |
|---|---|---|---|
| G1 | Market coverage | Every top-10 market by impressions or revenue has localized metadata House | App Analytics by territory; Play Console |
| G2 | Local keyword research | Each localized keyword field built from local research, not translated from English | Paid keyword data per storefront |
| G3 | Cross-localization (iOS) | Every extra localization indexed in your top storefront carries unique keywords | Cross-Localization Explorer |
| G4 | Localized creative | Top-5 non-English markets have localized screenshot captions at minimum House | Manual |
| G5 2026 | New languages reviewed | The 11 languages added March 31, 2026 checked against your traffic; decision documented; keyword fields filled to 100 characters | App Analytics by territory |
| G6 | Native QA | Each locale reviewed by a native speaker; no truncation or machine-translation errors | Manual |
The keyword field holds 100 characters, not 100 bytes. App Store Connect Help still says "up to 100 bytes", but Apple's product-page guide says "100 characters total", and saved strings back the guide. A developer on Apple's forums saved a 100-character Thai-and-Latin field through the API, while a longer one failed with "cannot be longer than 100 characters" (Apple Developer Forums, 2022), and a 100-character, 300-byte Thai field was accepted (Applyra, July 2026). So plan Tamil or Marathi fields in characters, not bytes. Vowel signs appear to count as characters too, so a full field shows fewer visible letters. To confirm it for your own locales, test the byte limit yourself on a draft version.
For G1 and G5, which languages to localize first walks through the traffic math.
8. 2026 discovery surfaces: are you using what's new?
This category checks whether you use three changes that added new places to win or lose the install. In June 2026, Apple announced header and search-result creative assets (Apple Newsroom, 2026). On October 29, 2025, it added keywords on up to 70 custom product pages (Apple Developer News, 2025). In May 2026, Google introduced Ask Play, building on AI Q&A that "already answers 95% of user queries" (Android Developers Blog, 2026).
| ID | Check | Pass threshold | How to check |
|---|---|---|---|
| H1 2026 | iOS 27 creative assets | Header and a chosen search-results asset ready for primary markets; 4+ suitable; no pricing, URLs or copyright symbols | App Store Connect Asset Library |
| H2 2026 | CPP coverage | At least 1 custom product page per major intent or paid channel House, each mapped to a traffic source (limit 70) | App Store Connect |
| H3 2026 | CPP keyword assignment | High-intent keywords assigned to the matching page; each keyword combination unique to one page | App Store Connect |
| H4 | Play custom store listings | Listings for keyword clusters, countries or lapsed users where intent differs (limit 50; one per country) | Play Console |
| H5 | Events and promotions | At least 1 In-App Event (iOS) or Play promotional content per quarter House | App Store Connect; Play Console |
| H6 2026 | Ask Play readiness | Play description opens with what the app does and for whom; listing and website answer the top 5 pre-install questions House | Manual: ask the questions, read the answers |
| H7 2026 | Brand defense in Apple Ads | Brand campaign live in top markets; brand results page checked on a current iOS device | Apple Ads |
On H1: as of September 24, 2026, Apple's What's new for developers page still labels creative assets and the Asset Library "coming this fall". Until then, H1 passes on assets ready to submit. Without a search asset, "your In-App Events, app previews, and screenshots will appear" instead (Apple Developer, Asset best practices, 2026). The header template is 3840×1646 px (about 21:9, not 3:2); full specs are in our iOS 27 creative assets guide.
For H3, Apple's rule is that "each keyword combination is unique to a single product page" (Apple Developer, 2026). So first decide which traffic deserves its own custom product page. On Play, each country takes only one of your 50 custom store listings (Google Play Console Help, 2026).
For H6, AppTweak's June 2026 analysis found that Ask Play draws on your Play description, then your website. As AppTweak puts it, "your store listing and your website now need to tell the same story" (AppTweak, 2026). More in how Ask Play changes Google Play ASO.
For H7: since March 2026, Apple Ads can place an ad at the top of results "or further down the page" (Apple Ads Help, 2026; MacRumors, 2026). That changes what brand defense buys you.
What does an audit finding look like?
A finding is a failed check with evidence, a point cost, a fix and an effort estimate. That format turns an audit into a roadmap. The card below is illustrative: the listing is invented and carries no outcome numbers.
Illustrative example · not client data
C3, C4, C5 · App Store text metadata
- Finding
- An invented photo-editing app, "[Brand]: Photo Editor", uses this keyword field:
photo, editor, filters, collage, free, retouch, blur, effect, effects, pic, selfie - Evidence
- 82 of 100 characters used, 10 of them spaces. "photo" and "editor" repeat the app name, "effect"/"effects" is a plural pair, "free" is filler.
- Points lost
- C3 0 of 2, C4 0 of 2, C5 1 of 2: 5 of 18 category points, or 4.2 of the category's 15.
- Fix
- Strip the spaces and four wasted terms. The cleaned field,
filters,collage,retouch,blur,effect,pic,selfie, is 46 characters, freeing 54 for researched terms. - Effort
- S. One metadata change, shipped with the next app version.
- Signal to watch
- B5: each new term returns the app in the top 200 within about 14 days of release. Then B3 and B4 at the next monthly review.
Rank findings in two passes: gates first, then points lost ÷ effort. Prefer to hand the scoring over? The fixed-price ASO audit covers the same ground with paid keyword and benchmark data, and ends in a prioritized 90-day roadmap.
Which ASO mistakes cost the most points?
The costliest mistakes trip a gate, because a gate caps the grade however good the rest is. A rating under 4.0 alone puts you outside the band where about 95% of featured App Store apps sit (AppTweak, 2026). For process errors rather than listing fails, see the strategy mistakes: set-and-forget, testing two things at once.
- Letting the rating slip under 4.0 in a top market (F1, a gate).
- Keyword lists or "#1", "Best", "Free" in Play metadata (D4, a gate with suspension risk).
- Wasting the keyword field: spaces after commas, repeated title words, plural pairs (C3, C4).
- Translating keywords instead of researching each market, and sizing new-language keyword fields in bytes instead of characters (G2, G5).
- Treating screenshot captions as a keyword field. After 2025 reports that Apple had started reading caption text for search, ConsultMyApp tested 64 screenshot phrases across 8 category-leading apps. Of those, 36 didn't rank, 27 were explained by existing metadata, and 1 was a possible anomaly (ConsultMyApp, June 2025). Apple hasn't confirmed caption indexing and the evidence is contested at best, so write captions for people.
- Reading Play trends across the 2026 metric change (A4).
- Sending every searcher to the default page: no CPP keywords, no custom store listings (H2–H4).
- Skipping the iOS 27 search asset, which leaves Apple's fallback to pick your search result (H1).
The ASO audit scorecard
Here's the whole rubric on one page. Score each check 0, 1 or 2, total each category, then apply the gates. No email, no form. Re-run it quarterly with the same thresholds so this quarter's 72 means what last quarter's did.
| ID | Check | Store | Run with | Score |
|---|---|---|---|---|
| 1. Foundations · weight 5 · max 12 | ||||
| A1 | Baseline exported | Both | Console | |
| A2 | Organic vs paid separated | Both | Console | |
| A3 | Change log | Both | Manual | |
| A4 | Play metric break | Play | Console | |
| A5 | Competitor set | Both | Manual | |
| A6 | Owner and cadence | Both | Manual | |
| Subtotal ÷ 12 × 5 | ||||
| 2. Keywords & rankings · weight 15 · max 16 | ||||
| B1 | Keyword universe | Both | Paid | |
| B2 | Brand keyword #1 (gate) | Both | Free tool | |
| B3 | Core terms in top 10 | Both | Free tool | |
| B4 | Top-50 coverage | Both | Free tool | |
| B5 | Indexing verified | iOS | Free tool | |
| B6 | Winnability | Both | Paid | |
| B7 | Paid-to-organic loop | iOS | Console | |
| B8 | Review language mined | Both | Free tool | |
| Subtotal ÷ 16 × 15 | ||||
| 3. App Store text · weight 15 · max 18 | ||||
| C1 | App name | iOS | Free tool | |
| C2 | Subtitle | iOS | Free tool | |
| C3 | Keyword field fill | iOS | Free tool | |
| C4 | No duplication | iOS | Free tool | |
| C5 | No wasted or risky terms | iOS | Free tool | |
| C6 | Promotional text | iOS | Console | |
| C7 | Description above the fold | iOS | Free tool | |
| C8 | Category | iOS | Manual | |
| C9 | Promoted in-app purchases | iOS | Console | |
| Subtotal ÷ 18 × 15 | ||||
| 4. Google Play listing · weight 10 · max 14 | ||||
| D1 | Title | Play | Free tool | |
| D2 | Short description | Play | Free tool | |
| D3 | Description density | Play | Free tool | |
| D4 | Policy-clean metadata (gate) | Play | Manual | |
| D5 | Feature graphic and video | Play | Console | |
| D6 | Screenshots | Play | Free tool | |
| D7 | Category and tags | Play | Console | |
| Subtotal ÷ 14 × 10 | ||||
| 5. Creative & conversion · weight 20 · max 20 | ||||
| E1 | Icon | Both | Free tool | |
| E2 | Search-visible frames | Both | Free tool | |
| E3 | Captions | Both | Manual | |
| E4 | Slot usage | iOS | Console | |
| E5 | App preview | iOS | Console | |
| E6 | Conversion vs peers | iOS | Console | |
| E7 | Search-to-page leak | Both | Console | |
| E8 | Testing cadence | Both | Console | |
| E9 | Device coverage | Both | Console | |
| E10 | Freshness | Both | Manual | |
| Subtotal ÷ 20 × 20 | ||||
| 6. Ratings & quality · weight 15 · max 14 | ||||
| F1 | Rating floor (gate) | Both | Console | |
| F2 | Rating trend | Both | Free tool | |
| F3 | Prompt design | Both | Manual | |
| F4 | Response coverage | Both | Console | |
| F5 | Complaint themes | Both | Free tool | |
| F6 | Android vitals (gate) | Play | Console | |
| F7 | Review integrity | Both | Manual | |
| Subtotal ÷ 14 × 15 | ||||
| 7. Localization · weight 10 · max 12 | ||||
| G1 | Market coverage | Both | Console | |
| G2 | Local keyword research | Both | Paid | |
| G3 | Cross-localization | iOS | Free tool | |
| G4 | Localized creative | Both | Manual | |
| G5 | New languages reviewed | iOS | Console | |
| G6 | Native QA | Both | Manual | |
| Subtotal ÷ 12 × 10 | ||||
| 8. 2026 surfaces · weight 10 · max 14 | ||||
| H1 | iOS 27 creative assets | iOS | Console | |
| H2 | CPP coverage | iOS | Console | |
| H3 | CPP keyword assignment | iOS | Console | |
| H4 | Play custom store listings | Play | Console | |
| H5 | Events and promotions | Both | Console | |
| H6 | Ask Play readiness | Play | Manual | |
| H7 | Brand defense in Apple Ads | iOS | Console | |
| Subtotal ÷ 14 × 10 | ||||
| Total (out of 100; single-store apps: ÷ applicable weight × 100) | ||||
| Gates failed (B2, D4, F1, F6). Any failure caps the grade at C | ||||
| Grade: A 85–100 · B 70–84 · C 50–69 · D below 50 | ||||
Frequently asked questions
What is an ASO audit?
An ASO audit is a structured review of your store presence, from keywords and metadata to creative, ratings, localization and new discovery surfaces, each scored against a pass threshold. The output is a ranked fix list. Our rubric uses 60 checks weighted to 100 points; 57 need no paid tool.
What should an ASO audit checklist include?
Both stores' metadata and creative, keyword rankings, ratings and app quality, localization, and the 2026 surfaces: iOS 27 creative assets, keywords on custom product pages and Ask Play. Every check needs a threshold, such as a Google Play crash rate below 1.09% (Android Developers, 2026).
How often should you run an ASO audit?
Quarterly, plus before major releases and after any listing change. Listings move fast: in 2025, 34% of top apps on both stores updated their description twice or more (AppTweak, 2026). Re-running the same rubric keeps scores comparable, so you can see whether fixes held.
Can you do an ASO audit for free?
Mostly. In our rubric, 57 of 60 checks run on free tools, App Store Connect, Play Console or a manual review. Only keyword universe, winnability and local keyword research need paid data. Automated score tools grade your public listing, not your own traffic data.
What's the difference between an ASO audit and an ASO strategy?
The audit is a scored snapshot of where your listing stands today. The strategy is the recurring loop of goals, research, optimization and testing that the audit feeds. Since almost 65% of App Store downloads follow a search (Apple, 2022 data), most strategies start with the visibility gaps.
The bottom line
- Fix the gates first: brand #1, Play policy, a 4.0+ rating and Android vitals.
- Don't skip the 2026 layer: 8 checks, from iOS 27 creative assets and CPP keywords to Ask Play and wake-lock vitals.
- Re-run it quarterly with the same thresholds, so scores stay comparable.
Score your listing with the scorecard first. Want it run with paid keyword data and a ranked 90-day roadmap? Book a free 30-minute call or start with the fixed-price ASO audit.
