app-store-optimization

Files (1)

• SKILL.md

SKILL.md (first lines)

## When to Use

Use when improving an app's visibility and conversion in the Apple App Store or Google Play Store: keyword research, competitor ranking analysis, metadata optimization (title/subtitle/keywords/descriptions), screenshot and A/B testing, review mining, launch checklists, or tracking ranking changes. Works for iOS and Android — always note which store each finding applies to.

## Workflow

### 1. Keyword Research
1. Pull current ranking keywords from first-party sources first:
   - **iOS:** App Store Connect → Analytics → App Store Search (impressions, taps, conversion by query). Apple Search Ads → Search Terms report for paid-converting terms.
   - **Android:** Google Play Console → Reports → Search terms report (impressions, store listing visits, installs, conversion rate per query).
   - **Search Ads API (iOS):** keyword suggestions endpoint — `GET /v4/campaigns/{campaignId}/adgroups/{adGroupId}/keywords/suggestions?query={term}&limit=10` with `Authorization: Bearer {token}` and `X-AP-Context: orgId` headers (v4/v5). Returns suggested keywords with relevance/matching fields — free discovery of terms users actually search.
2. Enrich with third-party tools (Sensor Tower / AppTweak / App Radar style fields): per-keyword **volume** (monthly search frequency), **difficulty/competition** (how hard to rank), **relevancy** (fit to the app), and **trend** (rising/falling). Without a paid tool, estimate volume from search-term reports and autocomplete frequency.
3. Score candidates: `score = volume × relevancy / difficulty`. Rank, keep the top 20-30; drop terms below your volume floor or relevancy < 0.5.
4. Bucket into head (high volume, hard), mid-tail (moderate volume, winnable), long-tail (low volume, high intent). Target mid-tail for the fastest ranking wins.

### 2. Competitor Ranking Analysis
1. Pick 3-5 competitors (direct + category leaders). Record each one's: title, subtitle, keyword field contents (iOS), description, current rank for your target keywords, rating score and count.
2. Build a keyword × competitor matrix: for each target keyword, note which competitors rank top-10 and which metadata terms win it.
3. Identify gaps: keywords competitors rank for that you don't (highest-ROI additions) and their weaknesses (missing keywords, weak ratings, stale screenshots).

### 3. Metadata Optimization
- **iOS:** Title (30 chars), Subtitle (30), Keyword field (100 chars, comma-separated — no repeats, no spaces after commas, don't repeat title/subtitle words). Front-load the title with the highest-value keyword.
- **Android:** Title (30), Short description (80), Full description (4000), tags. Google indexes the full description; repeat primary keywords naturally 2-3 times, no stuffing.
- Rule: every word in the keyword field must earn its place — remove anything not in your scored top-30. Re-audit quarterly or after major updates.

### 4. Screenshots & A/B Testing
1. Lead with the first 2-3 screenshots — they drive conversion. Show the core value proposition in 5-7 seconds; no feature-list dumps.
2. **iOS:** App Store Connect → Product Page Optimization — up to 3 treatments, 90-day test window; test title, subtitle, screenshots, or icon.
3. **Android:** Google Play Console → Experiments — test listing graphics/descriptions with store-provided traffic.
4. Test one variable at a time; run to statistical significance (p < 0.05; typically 2-4 weeks at your traffic level) before promoting a winner.

### 5. Review Mining
1. Pull all reviews (App Store Connect / Play Console exports, or AppTweak/Sensor Tower summaries).
2. Tag by theme: feature requests, bugs/crashes, pricing complaints, praise. Extract the exact phrases users use — they are keyword and description copy gold.
3. Prioritize by volume; feed the top 3 recurring themes into the next release/description update. Reply to 1- and 2-star reviews within 48h (rating impact + ranking signal).

### 6. Ranking Tracking
Re-check target keyword positions weekly (Sensor Tower/AppTweak keyword rankings, or manual App Store/Play search). Log movement and correlate with metadata/version changes to learn what moved the needle.

## Pitfalls
- Keyword-stuffing the title/keyword field (repeats, irrelevant terms) — Apple/Google can suppress rankings and it wastes the 100-char budget.
- Ignoring localization: keywords behave differently per locale; optimize each storefront separately, never translate literally.
- Chasing head terms you can't win: a top-50 rank on a huge term beats rank 200; win mid-tail first.
- Treating screenshots as an afterthought — conversion rate (impression → install) multiplies every ranking gain.
- Trimming the iOS keyword field without re-verifying: the field is evaluated as a whole; after changes, re-check rankings within 2-4 weeks.

## Verification
- Confirm metadata changes are live in App Store Connect / Play Console and respect character limits (30/30/100 iOS; 30/80/4000 Android).
- Export search-terms reports before and after the change; verify target keywords' impressions and conversion improved or held after 2-4 weeks.
- A/B tests produced a statistically significant winner with a documented lift (e.g., +8% conversion) before promotion.
- Review replies logged; theme counts captured and the top 3 themes fed into the next iteration.

From lukemcqueen/hermes-cortex, MIT.

More from lukemcqueen/hermes-cortex