App Store Optimizer

This skill handles App Store Optimization (ASO) tasks: keyword research, metadata writing, visual asset strategy, competitive analysis, localization, and A/B testing planning for iOS and Android apps.

Core Workflow

Step 1: Audit the Current Listing

Before making recommendations, collect the following inputs from the user:

• App name, category, and target platform (iOS / Android / both)
• Current app title, subtitle/short description, and long description
• Primary target audience and geographic markets
• Top 3–5 direct competitors (ask if not provided)
• Current ratings, review volume, and any known conversion data

Validation checkpoint: If inputs are incomplete, ask only for what is missing before proceeding.

Step 2: Keyword Research and Gap Analysis

1. Identify primary keywords (high search volume, high relevance to core use case).
2. Identify long-tail keywords (lower volume, higher intent, specific user problems).
3. Identify competitor keyword gaps (terms competitors rank for that the app does not).
4. Score each keyword on three axes: estimated search volume (H/M/L), competition level (H/M/L), relevance (1–10).

Example output format:

Validation checkpoint: Confirm keyword priorities with the user before writing metadata.

Step 3: Metadata Optimization

Write platform-specific metadata using the validated keyword list.

iOS Metadata

• Title (max 30 chars): Lead with the primary keyword. Example: Budget Tracker – Expense Log
• Subtitle (max 30 chars): Secondary keyword + key benefit. Example: Personal Finance & Money Goals
• Keyword field (max 100 chars): Comma-separated, no spaces after commas, no repeats from title/subtitle. Example: spending,bills,savings,weekly,cash,envelope,planner
• Long description: Open with a one-sentence value proposition. Follow with a bulleted feature list. Close with social proof (rating, downloads) and a CTA. Integrate keywords naturally.

Android (Google Play) Metadata

• Title (max 30 chars): Primary keyword + brand or differentiator.
• Short description (max 80 chars): Hook + primary value prop. Indexable by Play Store search.
• Long description (max 4,000 chars): First 167 characters are shown before "Read more" fold — make them count. Structure: hook → features (bulleted) → social proof → CTA. Google indexes the full text, so integrate all target keywords naturally.

Validation checkpoint: Run a self-check — confirm no keyword appears in both the keyword field and title/subtitle (iOS), and that the long description's opening is compelling before truncation.

Step 4: Visual Asset Recommendations

Provide specific, actionable direction for each asset type.

App Icon

• Must be recognizable at 16×16px (small search result size).
• Recommend a dominant color that contrasts with category competitors (provide a specific suggestion based on the category).
• Suggest icon concept: symbol vs. lettermark, minimal vs. detailed, with rationale.
• Propose 2–3 A/B test variants with clear hypotheses. Example: "Variant A: green piggy bank (category-conventional) vs. Variant B: white upward arrow on dark navy (differentiator) — test whether category familiarity or distinctiveness drives higher tap rate."

Screenshot Sequence (iOS: up to 10; Android: up to 8)

• Screenshot 1 (Hero): State the core value proposition in a headline overlay. Show the most compelling UI state. Example headline: "Know exactly where your money goes."
• Screenshots 2–3 (Core features): Walk through the 2 highest-value features. Each screenshot = one feature + one benefit statement.
• Screenshots 4–5 (Depth/social proof): Secondary features, awards, review quotes, or download milestones.
• Recommend device frame style (with frame vs. frameless) based on current category conventions.

App Preview Video

• Duration: 15–30 seconds (iOS), up to 30 seconds (Android).
• Structure: 0–3s hook (problem or surprising moment) → 3–20s feature demo (real UI, no mockups) → 20–30s CTA + value reinforcement.

Step 5: Localization Strategy

For each priority market beyond the base locale:

1. Identify whether a full translation or cultural adaptation is needed — these are different scopes with different costs.
2. Flag screenshot text and imagery that contains cultural references requiring adaptation.
3. Note markets where competitors are weak — these are higher-ROI localization targets.
4. Prioritize: Tier 1 (translate everything including screenshots), Tier 2 (translate metadata only), Tier 3 (monitor only).

Step 6: Review and Rating Strategy

• Identify the optimal in-app prompt timing: after a positive user action, not on first launch or after a failure.
• Flag review velocity patterns that could indicate a rating drop risk.
• Flag if any requested tactic conflicts with platform policies (e.g., incentivized reviews violate both Apple and Google policies).

Step 7: A/B Testing Roadmap

Sequence tests to isolate variables. Only one element should change per test.

Step 8: Performance Monitoring

Define a cadence for ongoing tracking:

• Daily: Keyword rankings for top 10 primary keywords, download volume, rating average.
• Weekly: Conversion rate (impressions → product page views → installs), search visibility score, review sentiment.
• Monthly: Full competitive re-audit, keyword gap refresh, strategy adjustment based on trends.

Report format for weekly check-in:

Week of [DATE]
- Organic downloads: [X] (+/-% vs. prior week)
- Conversion rate: [X%] (+/-% vs. prior week)
- Top keyword movement: [keyword] moved from #[X] to #[Y]
- Action item: [specific change to make this week]

Platform-Specific Rules

Communication Style

• Lead with data and specific numbers, not generalizations.
• When making a recommendation, state the rationale and expected outcome. Example: "Move 'expense tracker' to the app title (from the keyword field) — title keywords receive 3× the ranking weight of keyword-field terms on iOS."
• When a user's request is platform-agnostic, ask which platform or provide separate recommendations for each.
• Flag when a requested optimization conflicts with platform guidelines (e.g., incentivized reviews violate both Apple and Google policies).