Trends by Install Cohort
The Trends by Install dashboard provides visibility into how install cohorts perform over time. By tracking KPIs cohort by cohort, the dashboard helps identify whether changes in segment performance are driven by user quality (i.e. monetisation, retention) or acquisition dynamics such as CPI shifts (e.g. creative changes, competitor landscape).
What is in this dashboard
6 tiles: ROAS, LTV, ARPDAU, Retention, CPI, Spend
X-axis: Install cohort date (each point = all users who installed in that period)
Y-axis: KPI shown in the tile title
Solid line = actuals; Dashed line = predictions
How it works
ROAS, LTV, ARPDAU, Retention
Tracked at D1, target DSI, and optimization DSI
Displayed as actuals (solid line) where data exists; predictions (dashed lines) fill gaps where data has not yet matured
Example: In weekly / monthly cohorts, earlier install cohorts within the date range may have up to D30 actuals, while later days only have D23 actuals; values beyond D23 will be predicted
CPI, Spend
Always actuals, as they are realised at the date of install
The dashboard is most useful for:
Conducting deep dives into specific segments (e.g platform, campaign, geo)
Understanding whether ROAS shifts come from:
LTV / retention changes (e.g. due to creatives, monetization design, user engagement quality)
CPI changes (e.g. new creative, budget allocation, competitor activity)
Broader factors such as seasonality or macroeconomic shifts
Providing context for segment-level changes by tying performance to install cohorts
Available filters and controls:
Paid vs Blended (include / exclude organic uplift)
Net vs Gross (with / without platform fees / taxes)
Pivot dimensions:
By default = DSI
If DSI is pivoted, no second pivot can be added
If DSI is not pivoted, up to 2 dimensions may be selected (Campaign, Network, Geo, Platform) in any combination
Install groupings: Cohorts may be grouped by daily, weekly, or monthly installs.
Date range selector: Include cohort data of the last 7, 14, 28, 60 days, or 3 months.
Segment filters: Views can be narrowed by Geo, Platform, Network, or Campaign, either individually or in any combination.
DSI selector: Controls which Dx is shown when DSI is not the pivot.
Tips and warnings
✅ Start broad (e.g. weekly cohorts across all networks) to spot high-level shifts; then filter down (e.g. campaign-level on iOS) to isolate the driver
✅ Compare CPI, LTV, and ROAS tiles together to separate acquisition-driven changes from monetization driven ones
⚠️ Smaller cohorts (daily or narrow filters) can show volatile lines due to low data volume
⚠️ Predicted segments (dashed lines) are more uncertain for later DSIs - interpret trends directionally, not as exact values