agentsclimarketplace

Aso ab testing

Skill FelixGraeber/claude-aso-audit-skill/skills/aso-ab-testing

ASO audit skill pack for Claude Code: audit App Store and Google Play listings, keywords, metadata, visuals, reviews, and competitors.

Install
npx -y skills add FelixGraeber/claude-aso-audit-skill --skill aso-ab-testing

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

A/B testing strategy for iOS Product Page Optimization (PPO) and Android Store Listing Experiments. Hypothesis design, variant creation, statistical significance guidance. Triggers on: "A/B test", "experiment", "PPO", "product page optimization", "store listing experiments".

SKILL.md

3.0 KB, as published. Nobody here has run it

ASO A/B Testing — Experimentation

Capabilities

  1. Test hypothesis generation from audit findings
  2. Platform-specific experiment design
  3. Variant recommendations (control vs treatment)
  4. Statistical significance and duration guidance
  5. Results interpretation framework
  6. Sequential test roadmap planning

Platform Capabilities

iOS: Product Page Optimization (PPO)

AspectSpecification
Testable elementsApp icon, screenshots, app preview video
NOT testableTitle, subtitle, keywords, description
Max treatments3 (plus original)
Traffic splitApple-controlled
Min duration7 days recommended
Max duration90 days
AudienceAll users or specific locales
Active tests1 at a time (on default product page)

Android: Store Listing Experiments

AspectSpecification
Testable elementsIcon, feature graphic, screenshots, short description, full description, promo video
Max experiments5 localized + 1 main simultaneously
Traffic splitConfigurable
Min duration7 days recommended
AudienceDefault or country-specific listings

Test Design Framework

1. Hypothesis

State: "Changing [element] from [current] to [proposed] will [increase/decrease] [metric] because [reason]."

2. Element Selection (priority order)

  1. Screenshots (highest conversion impact, testable on both platforms)
  2. App icon (affects browse + search impressions)
  3. Short description / feature graphic (Android only for text)
  4. Preview video (presence vs absence)

3. Variant Design

  • Change ONE element per test (isolate variable)
  • Make the change meaningful (not subtle)
  • Have clear visual/copy difference between control and treatment

4. Duration & Sample Size

  • Minimum 7 days (capture weekday + weekend patterns)
  • Need 90%+ confidence level
  • Rule of thumb: ~1000 page views per variant for meaningful results
  • Account for seasonal effects

5. Success Metrics

  • Primary: Install conversion rate (page view → install)
  • Secondary: First-time installers, 1-day retention (Android)

Output Format

# A/B Test Plan: [App Name]

## Test 1: [Element Being Tested]
- Hypothesis: [statement]
- Platform: iOS PPO / Android Experiment
- Control: [current element description]
- Treatment: [proposed change]
- Expected impact: [conversion increase estimate]
- Duration: [recommended days]
- Success criteria: [metric + threshold]

## Test Roadmap (Sequential)
1. [highest impact test first]
2. [second test]
3. [third test]

Available Tools

Read, Bash, Write, Glob, Grep

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.