Cohort dashboard

At a glance: The Cohort dashboard groups app users by acquisition date. Perform performance analysis across user life-cycles. Use the dashboard to gain insights as to how campaign differences alter campaign metrics. 

Revenue by days post-install broken down by media source 

Full_cohort2_us-en.png

Cohort dashboard

To use Cohort, select users from one or more apps and segment them by conversion time. Cohort metrics like revenue, ROI, and event conversion rates are available. Break the cohort down, for comparison purposes, into dimensions like campaign or media source. Doing so can uncover patterns or changes in performance over user or campaign life-cycles. See Cohort use cases

Tip

Want to know more about cohort reports? Check out this short, informative course on the AppsFlyer Learning Portal.

Cohort principles

OV.png
  • Cohort type
    • Unified
      • The combined performance of user acquisition and retargeting campaigns (including cost and revenue).
      • If an event is attributed to both conversion types, only the retargeting event is included. Meaning, the report is generated using the actual last touch media source. If you are familiar with raw data, it means events where is_primary_attribution=true
    • User acquisition: performance data from installs (reinstalls excluded unless specially enabled)
    • Retargeting: performance data from re-attributions or re-engagements
  •  Trend type
    • User lifetime performance: Users are grouped by conversion time. This enables you to compare user performance post-conversion, starting with the conversion day, referred to as cohort day 0. Cohort disregards the event hour and uses only the event day. This trend is typically used to evaluate user performance relative to the media source.
      • For hour granularity the maximum number of hours is 72.
      • For day granularity the maximum number of days is 180. 
      • For week granularity the maximum number of weeks is 156
      • For month granularity the maximum number of months is 36
    • KPI by attribution time: Users are grouped by conversion time. This enables you to evaluate campaign performance over time and to view campaign KPIs relative to the conversion time.
  • Cost-related metrics: In Cohort cost data is included only if there is an associated attribution (install, re-engagement, or re-attribution). In other words, if a campaign incurs cost based on clicks and impressions but has no attributions the cost isn't included in cohort reports. This principle holds true for all levels of the advertising hierarchy. 
  • Granularity options
    Granularity type  Permitted period Remarks
    Days Select 1–60 days  
    Hours Select 1–72 hours out of the previous 72 hours Cost related metrics not available
    Weeks Up to 26 weeks  
    Months Up to 12 months  

Example: User lifetime performance trend vs. KPI by attribution time trend

In this example, the same raw data is used to generate the two cohort trend types. The raw data table contains attribution events related to different users A, B, C, and D. The users engage with an ad, install the app, and post-conversion make one or more purchases. Cohort day 0 is the install day.

CohortExample_en-us.jpg

Metrics available

    Functions
Always/ Optional 

Metric

Count cvr (conversion rate) Rate Sum Unique users eCPA
Number Percentage Percentage Number Number Number
Always Users (4) - - - - -
Always Cost (3) - - - - -
Always Average eCPI - - - - -
Optional "event_name"   - (5)
Optional Revenue - - - -
Optional ROAS (3) - - - - -
Optional ROI (3) - - - - -
Optional Sessions (2) - - (1) -
Optional Uninstalls (3) - - - -

(1) Unique sessions returns when aggregation_type=on_day
(2) Retargeting (re-engagement and re-attribution) sessions are available for sessions that occur starting from July 13, 2020. This means that sessions occurring until July 12, 2020, do not display. 
(3) Not available in the following cases:

  • View type is unified
  • Granularity is hours. Calculated daily using UTC time. 

(4) The Users metric counts the number of attributions (installs, re-engagements, re-attributions) taking place during the selected period.   

(5) Effective cost per action (eCPA): Cost / Cumulative number of unique users performing the event.

Using Cohort

Setting up an analysis

  1. Go to Dashboard > Cohort.
    The Cohort dashboard opens. The granularity and conversion date range display.
    mceclip0__1_.png
  2. Select the granularity. 
  3. Set the attribution period (user acquisition or retargeting engagement). Data relating to the most recent dates can result in partial data indicated by dashes in charts and tables.
  4. Click Edit Cohort.
    The Cohort setup window opens.

    CohortSetup.jpg

  5. Complete the following:
    • Cohort type:
      • User acquisition.
      • Retargeting
      • Unified: includes the in-app events of both user attribution and retargeting attributions. Note: retargeting events take precedence over user acquisition events meaning the report contains events where is_primary_attribution=true
    • Select a Trend type:
      • User lifetime performance. 
      • KPI by attribution time.
    • Users of the apps: Select 1–many or all (means the top-100 apps in your account). Note: Agencies can only select one app.
    • Grouped by: Select one or more dimensions [Default: Media Source]. 
    • Minimum cohort size: Set the minimum cohort size. This avoids cluttering your report with insignificant (small) cohorts. [Default=1]
    • Optional fields:
      • Filtered by: Select one or more dimensions to filter by. There are no default filters.
  6. Click Apply changes.
    The report displays.
  7. To complete the cohort setup, complete one of the trend-type procedures that follow. 
  8. [Optional] Managing saved cohort reports.

Trend type—User lifetime performance

Users are grouped into cohorts by post-conversion days. The conversion day is referred to as post-conversion day 0. The next as post-conversion day 1 and so on.

  • For day granularity the maximum number of days is 180. 
  • For week granularity the maximum number of weeks is 156
  • For month granularity the maximum number of months is 36

DaysPost.jpg

  • Select a metric, function, and per user
    Note! Not all combinations of metric, function, and per-user are available. 
    • Metrics: Revenue,Sessions, Uninstalls, ROI, In-app event, ROAS.
    • Functions: The metric is calculated using functions. The function list varies depends on the metric.
      • Sum: the total revenue on a cohort day. 
      • Count: count of events. For example, the number of sessions, number of revenue events, number of uninstalls, etc. 
      • Unique users: the unique number of users that performed an event. That means, for example, that if a user opened the app 10 times on a specific day that the result shown will be 1.
      • In-app conversion: conversion rate from install/retargeting to an in-app event. The number of unique users who perform a specific in-app event for the first time ever divided by the number of unique users in the cohort. For example, the number of [unique] users in a cohort is 100. Of these, 20 perform a specific in-app event for the first time. The In-app conversion rate is 20/100=20%. 
    • Per-user: when selected, the sum or count is divided by the number of app users.
  • Display duration: The number of cohort days to display post-conversion. Day 0 is the conversion day, Day 1 the day following conversion, and so on. Note: The display duration is truncated if the result is a future date.
  • Table or Chart: 
    • Charts display the top 5 or 10 ranked by the number of app users in the cohort. This is irrespective of the dimension displayed.
    • Tables display up to 50 rows. The users, cost, and average ECPI (effective cost per install) always display irrespective of the metric selected.
  • On day or Cumulative:
    • On day means the value of the metric for that day only.
    • On day doesn't provide Cost related metrics.
    • Cumulative means the cumulative or running total of the metric. Note! In the context of a unique users metric, it means the number of unique users having performed a given event until a given cohort day. It is not the sum of unique users of the preceding days. 

Trend type—KPI by attribution time

Users are grouped into cohorts by attribution time

KPIdays.jpg

  • Select a metric, function, KPI days, and per user
    Note! Not all combinations of metric, function and per-user are available. 
    • Metrics: Revenue, Sessions (not available in unified view), Uninstalls, ROI, In-app event, ROAS.
    • Functions: The metric is calculated using functions. The function list varies depends on the metric.
      • Sum: the total revenue on a cohort day. 
      • Count
      • Unique users: the unique number of users that performed an event. That means, for example, that if a user opened the app 10 times on a specific day that the result shown will be 1.
      • In-app Conversion: conversion rate from install/retargeting to an in-app event. The number of unique users who perform a specific in-app event for the first time ever divided by the number of unique users in the cohort. For example, the number of [unique] users in a cohort is 100. Of these, 20 perform a specific in-app event for the first time. The In-app Conversion rate is 20/100=20%. 
    • KPI time: Select from Day 0, Day 3, Day 7, etc. Day 0 is the conversion day only. Day 3 means the conversion day and the next three days, a total of four days, likewise, for the remaining days. The KPI value is totaled for the period selected. 
    • Per-user: when selected, the sum or count is divided by the number of app users.
    • Table or Chart: Charts display the top 5 or top 10 ranked by the number of app users in the cohort. This is irrespective of the dimension displayed in the chart. Tables display up to 50 rows.
    • On day: The value only on the cohort Day specified. For example, In the case of Day 7 this means the value on post-conversion day 7.
    • Cumulative where: The combined value of all the days specified in Days. For example, in the case of Day 7, this means the total of the conversion day and the following 7 days. Note! In the context of unique users, it means the cumulative number of unique users.

List of dimensions for grouping and filtering

Dimension Comments Group by Filter by
Ad  
Ad ID  
Adset  
Adset ID  
Agency  
Apps Relevant when the report contains multiple apps x
Attributed touch-type Values: Click, impression, or NA (organic)
Campaign  
Campaign ID  
Channel  
Conversion type

Cohort type dependant:

  • User acquisition: Install
  • Retargeting: Re-engagement, Re-attribution
Country (Geo) Based on user IP address

Time. In the case of:

  • Daily granularity is the date
  • Hourly granularity is the hour of the day
  • Weekly granularity is the week of the year
  • Monthly granularity is the month of the year

Cohort type dependant:

  • User acquisition and unified: The time the app was launched for the first time.
  • Retargeting: The time the app was launched for the first time after re-engagement.

 

x
Install app store

Relevant for apps implementing Android multi-store attribution. Reach out to your CSM to enable this dimension. 

Keywords

Search keywords provided by the integrated partners

Media source  
Revenue Type

 Regular, Ad monetization

 x
Site Id  
Sub Param 1  

Using My reports—Save and select reports

  • My reports enable you to save the current dashboard settings for future use. 
  • Select a saved report to display the most current cohort data available.
  • You can save one or more reports.
  • Only you can access your My reports. 
Feature

Description

My reports

MyReports.png

Select a report to display the most current data available.

Save changes 

Save the report. Note: The currently selected report is overwritten. 

Action options (Elipse.png)

 

  • Save as new
    • Save the report using a different report name
    • [Option] Make this report default report
  • Rename: Change the report name
  • Set as default: Make the report the default report. When you open Cohort this is the report that displays initially
  • Delete: Delete the report

Partial data

Use the Real-time Cohort control to select the data display mode:

  • [Default] Real-time:
    • Attribution data updated continuously in real-time including revenue.
    • Cost data updates 6 times a day on average every four hours.
    • Partial data (indicated by dashes in reports and charts) is included
  • Daily: Data is updated daily and available a few minutes after midnight. Data of partial days does not display. 

What is partial data?

For a given day in the conversion data range, if the number of days that have elapsed since the conversion date is less than the cohort day, that cohort day contains partial data. Meaning partial data occurs on any day where the cohort day > current date-conversion date.

Export the cohort to a CSV file

To export cohort reports to CSV files:

  • In the upper-right side of the Cohort reports window, click Download.png(export).
    The cohort report CSV file is downloaded.

Information

Cohort for agencies

  • The advertiser needs to grant the agency permission to use Cohort. 
    • To grant permission to an agency, the advertiser:
      • Goes to Configuration > Active Integrations.
      • Select the agency.
      • Go to the permissions tab, enable Allow access to your retention and cohort reports.
  • Agencies are limited to accessing data of traffic generated by the agency. They can't see data related to organic traffic or non-organic traffic generated elsewhere.

Use cases

Revenue sum by country

Chart: Cumulative revenue per user

CohortGraph.png

Table: Revenue per day

CohortTable.png

In-app event conversion

An advertiser defines an in-app event af_purchase. The cohort chart displays the percentage of app users in the cohort who made a mobile purchase. We can see on day 10 for instance that close to 30% (11.11%+9.71%+7.47%) of the cohort users performed a mobile purchase

.Cohort-inapp

FAQ

Question Answer

Can I view Apple Search Ads keyword Cost and ROI?

Yes, you can.
Apple Search Ads provides us with the cost for each keyword per day broken down by Campaign/Adset/Ad. Note: Breakdown by Geo/country is not available.

For a given media source, why does the same campaign name display more than once?

  • If you use the same campaign name with campaigns having different campaign IDs then the name displays more than once. 
  • You should allocate campaign IDs unique names
  • When changing campaign names, data is aligned to the last name used.

Can I view organic users?

Organic users, by default, are included in Cohort reports.

To limit the Cohort report to organic users only:

  1. In the Cohort dashboard, click the Edit cohort button.
  2. In the Cohort setup window set Filter by > Media source is Organic:

    cohort_setup_filtered_by_media_source_is_organic.png

  3. Click Apply changes
Can I see data from deleted installs? 

Deleted installs, such as installs from SRNs that were deleted due to data retention policies or GDPR deleted installs, can not be seen in the cohort dashboard.

For more information click here.

Can I compare data for daily view to data in weekly or monthly views?

It’s not easy to compare data from a daily granularity to a weekly or monthly granularity, because the cohort LTVs don't align. 

When viewing cohorts in daily granularity, Day 0 displays data for the day of the installs. Day 1 is the day after the install, Day 2 is 2 days post-install, etc. 

When viewing cohorts in weekly granularity, data displays for all installs made in the entire Week 0. Week 1 is the week after the install, Week 2 is 2 weeks post-install, etc. 

When viewing cohorts in monthly granularity, data displays for all installs made in the entire Month 0. Month 1 is the month after the install, Month 2 is 2 months post-install, etc. 

This means the LTV timelines don’t align. For example, if you look at data for January 9-15: If you use daily granularity, Day 0 contains data for the first day of all installs made January 9-15. And Day 6 contains data for the first 6 days post-install of all installs made January 9-15. Meanwhile, in weekly granularity, Week 0 contains data for the entire period for all installs made January 9-15. So for installs on January 9, the data is like Day 6. But for installs on January 15, it’s like Day 0.

Traits and limitations

Trait Remarks 
Ad network access to Cohort Not available
Agency access

Available:

  • Limited to one app.
  • Hybrid accounts can't access Cohort.
  • Require access permission on an app-by-app basis.
  • Can't select multiple apps.
  • Access to organic data isn't possible.
Agency transparency Available starting January 10, 2021
App-specific time zone
  • Available
  • Exception: In all of the following cases, AppsFlyer ignores the app-specific time zone and uses the default time zone being UTC +0:
    • The cohort report contains data from multiple apps, and the apps have different time zones.
    • The apps all have the same time zone, and the time zone for at least one of the apps changed during the cohort report period. To determine the report period:
      • Start date: The first date in the conversion date range
      • End date: Is calculated by adding the number of days in the display duration to the last date in the conversion date range. Example: Last date: Jan 1, 2019. Report duration: 30 days. End date: Jan 31, 2019
    • Weekly and monthly data will be displayed in UTC timezone
App-specific currency 
  • Available
  • Exception: All apps need to use the same currency. If they don't, USD is used.
Limitations
  • CSV download files:
    • Maximum 10,000 rows.
    • Is not available for the KPI by attribution time trend.
  • Charts: display the top 5 or top 10 ordered by the number of users in the cohort.
  • Tables: Display the first 50 rows ordered by the number of users in the cohort.
  • Reinstalls: If post-reinstall evens are considered. unattributed organic, they aren't included in Cohort. If they're attributed to the first install, they are. Learn more
  • Cost: AppsFlyer does not report cost data for keywords that contain uppercase letters in the cohort report or cohort API.
  • Limit of 10 CSV exports per minute and 100 per day (per user).
  • Limit of 20 data queries per minute and 500 per day (per user). Changing time period, grouping, KPI, etc. is considered a new query.
Organic data Included
Non-organic data Included
Data freshness
  • Real-time
  • Exceptions:
    • Daily cohorts:
      • Cost, ad revenue, and uninstall metrics update Daily.
      • Uninstalls are calculated once a day using the UTC clock. 
    • Weekly and monthly cohorts:
      • Real-time data is not available. The data will only show after it has been completed (meaning at the end of the week or month).
        However, data, where some of the install data has been completed, such as some data has finished (post-install time) and others haven't started yet, will be displayed.
Historical data
  • Daily cohort: 2 years
  • Weekly/monthly cohorts: 3 years
Geo/country
  • Available
  • When geo is unknown (N/A), ad revenue data doesn't display when grouping by geo.
Session count

Session count for Day 0 in the Retargeting view might be lower than the user count.
There are a few reasons this can happen. The most common are:

  • The  re-engagement session has already been attributed before the re-engagement is recorded. In this case, it will not be attributed to retargeting session
  • Filtering by other settings such as the Minimum time between sessions. In this case, the re-engagement is recorded. However, the sessions are filtered and won’t be recorded
Retargeting and Unified data Monthly session data is only available from the first of September 2021 onwards
Ad revenue
  • Not available when viewing cohorts with hourly granularity.
  • There may be discrepancies between the ad revenue amounts displayed in the Cohort and Overview dashboards. This occurs when networks report revenue in UTC and the apps have a different timezone setting.
  • For af_ad_revenue events, the unique users metric isn't available when the aggregation type is:
    • "Cumulative".
    • "On day," for dates between October 5, 2022-February 16, 2023.
Week and month granularity
  • Week and month granularity is only available on the Cohort dashboard.
  • Unique users and conversion rate metrics are not available if the data includes installs from before November 2022.
  • Cost-related metrics are not available when grouping by Week or Month. These include:

    • Cost
    • eCPI
    • ROI
    • ROAS 
  • Historical data is only available from the year 2021 and on.
Cohort period
  • The conversion period is referred to as 0 (Hour 0, Day 0, Week 0, or Month 0). The next period after the conversion is referred to as 1 (Hour 1, Day 1, Week 1, or Month 1), and so on.
  • In AppsFlyer a cohort period doesn't take into account the specific install timestamp. Rather, cohort hours are rounded down to the nearest hour, and cohort days, weeks, and months are based on the calendar day, week, or month. This may cause discrepancies when comparing AppsFlyer cohort data to the cohort data of other networks, where all cohort periods are determined by the specific install timestamp (meaning an hour is 60 minutes post-install timestamp, a day is the 24-hour post-install period after the install timestamp, etc.).