Analytics

Key metrics and concepts used in the AttributeHQ analytics dashboard.

Key Metrics

CPI (Cost Per Install)

CPI = Total Ad Spend / Number of Paid Installs

Lower CPI means more efficient ad campaigns. Import cost data via the dashboard or API to calculate CPI.

ROAS (Return on Ad Spend)

ROAS = (Revenue from Paid Users / Ad Spend) × 100%

A ROAS of 150% means you earned $1.50 for every $1 spent. AttributeHQ calculates ROAS by linking revenue events to their attributed install source.

LTV (Lifetime Value)

LTV = Total Revenue / Number of Users

LTV is calculated per cohort and per media source, allowing you to compare which acquisition channels bring the most valuable users.

CVR (Conversion Rate)

CVR = (Installs / Clicks) × 100%

Measures how many ad clicks convert to installs. Low CVR might indicate poor ad targeting or app store listing issues.

Retention (D1, D7, D30)

D7 Retention = Users active on Day 7 / Users who installed on Day 0

Tracks what percentage of users return to your app after installation:

Cohort Analysis

A cohort is a group of users who installed your app on the same day. Cohort analysis tracks how these groups behave over time.

The retention heatmap in the dashboard shows:

• Rows: Install date (cohort)
• Columns: Days since install (D1, D7, D30)
• Cells: Retention percentage (color-coded)

This helps identify:

• Which campaigns bring users who stick around
• Whether product changes improve retention
• Seasonal patterns in user engagement

Revenue Analytics

Revenue is tracked via trackRevenue() in the SDKs. The analytics dashboard shows:

• Revenue over time — Daily/weekly/monthly revenue trend
• Revenue by source — Which ad networks drive the most revenue
• ARPU (Average Revenue Per User) — Revenue / total users
• LTV by cohort — How much each cohort generates over time

Multi-Touch Attribution Models

The analytics API supports four attribution models:

Compare models via the API:

# Last-touch (default)
curl ".../analytics/attribution?model=last-touch"
 
# Linear attribution
curl ".../analytics/attribution?model=linear"

Different models tell different stories:

• Last-touch shows which channels close the deal
• First-touch shows which channels create awareness
• Linear gives a balanced view across the journey

Data Sources

All analytics queries are powered by ClickHouse for sub-second response times, even with millions of events.