App Analytics
Interpret ASO and App Store performance data and turn it into decisions for metadata, creatives, pricing, reviews, releases, and experiments.
ASO.dev Measurements context
ASO.dev Measurements compares current vs previous periods and shows release-date markers on charts. Use those release markers to connect metric changes to metadata, screenshot, pricing, event, or version changes.
Common metrics:
- impressions, product page views, downloads, conversion
- in-app purchases, subscription/revenue metrics when available
- retention, sessions, active devices
- ratings, reviews, review velocity
- release dates and before/after periods
Workflow
- Identify the question: conversion drop, install growth, revenue change, retention, experiment result, review/rating movement, or release impact.
- Capture period, country, locale, source/channel, app version, and changed surfaces.
- Compare current period to previous period and mark releases/campaigns.
- Separate correlation from causation.
- Recommend the smallest next experiment or investigation.
Launch or major update readout
For a new launch or major update, inspect the first 14 days closely when data exists:
- compare daily trend against release markers and previous comparable period
- watch conversion, downloads, ratings/reviews, crashes/complaint themes if supplied, and revenue/subscription starts if relevant
- distinguish launch novelty, featuring/ad bursts, seasonality, and metadata/creative effects
- turn findings into the next metadata, screenshot, review, pricing, or CPP/PPO action
Output
# App Analytics Readout
App:
Market:
Period:
Comparison period:
Question:
## Metric changes
| Metric | Current | Previous | Change | Interpretation |
|---|---:|---:|---:|---|
## Timeline
| Date | Release/campaign/change | Metrics affected | Notes |
|---|---|---|---|
## Diagnosis
- Likely driver:
- Alternative explanations:
- Data gaps:
## Next actions
| Priority | Action | Skill/tool | Success metric |
|---|---|---|---|
Rules
- Do not invent metrics or attribution.
- Do not call a change causal without enough evidence.
- Segment by country, locale, traffic source, and version when possible.
- For subscription-specific churn, trial, billing, or win-back analysis, use
subscription-lifecycle.