[Beta] Predict dashboard

Premium
For:
Marketers
Last update:

beta_feature.png

At a glance: Use the Predict dashboard to obtain insights about the predicted LTV of a campaign's acquired users at its earliest stages. Explore and analyze users' predicted performance metrics (KPIs) and eCPI (effective Cost Per Install), and compare predicted results among user cohorts. Based on these insights, make decisions about setting campaign bids, and which campaigns to stop, boost, or optimize.

predict_dashboard.png

Overview

Predict uses predictive analytics to provide accurate LTV-based predictions of campaign success as soon as 48 hours post-install. The Predict dashboard includes 2 primary components:

  • An interactive bubble chart that shows the distribution of predicted KPIs for new users. Each bubble represents a group of users ("cohort") grouped by the characteristics you specify: media source, campaign, geo, site ID, adset, and more.
  • A full, sortable results table, incorporating additional data fields for each cohort.

Further investigation of specific data points allows you to extract early insights about the quality of your campaigns and helps answer questions such as:

  • Will your campaign be successful?
  • Which campaigns and media sources provide the best users?
  • Will your campaign generate good users, or should you reduce campaign costs? 

Based on these insights, you can determine how to set campaign bids and make decisions about which campaigns to continue, stop, boost, or optimize.

Opening the dashboard

To open the dashboard: Go to Labs > Predict. 

Dashboard components

The Predict dashboard is made up of two tabs, the Overview, and the Validation report tab.

Overview tab

The following sections describe the main components of the overview tab.

Filter bar and attribution method selector

predict_dashboard_filter_bar.png

Use the filters and attribution method selector to view the information most relevant to you.

Note: Selections apply to both the bubble chart and the results table.

Filter

Description
App The app for which data is displayed
Dates 

The dates for which data is displayed. Use the date picker within this filter to change the displayed date range.

Notes:

  • By default, the dashboard displays data:
    • for the past 7 days (when AF model attribution is selected)
    • for the last 7 days available (when SKAN or SSOT attribution is selected)
  • The data displayed is determined by install date.
    • When SKAN or SSOT attribution is selected, this is an estimated date that can vary by up to 2 days.
  • SKAN data is available 3 days after the install date
  • AF model attribution data will be available 2 days after the install date
  • Data is refreshed daily.

Attribution method selector

Select whether you want to view predictions based on:

  • SKAN attribution data (SKAN)
  • AppsFlyer attribution model (AF model)
  • Single Source of Truth (SSOT)

Install type

Select whether you want to view predictions based on:

  • Installs and redownloads (all install types)
  • Installs only
  • Redownloads only (as reported by the App Store)

Note: This filter is relevant only when SKAN attribution is selected.

Media source

The ad network attributed with the install

Note: For agency-driven traffic, the actual media source is displayed irrespective of the agency transparency status. Consequently, you may see media sources that are unfamiliar to you. 

Campaign

Campaign name and ID as set by the ad network 

Notes:

  • AppsFlyer translates the postback campaign ID to the one assigned by the ad network.
  • If a given campaign ID is associated with multiple campaigns, then all campaign names are displayed in the dashboard.
  • The ad network provides the campaign ID and name per postback. You can view data by campaign ID in aggregated reports, accessible via the API. 

Geo

Territory or country (based on data supplied by the ad network or the device's IP address)

Notes:

  • The Geo filter is not supported for SSOT attribution.
  • For SKAN attribution only: Starting with iOS 14.6, the device's IP address is masked by an Apple proxy server. This means that the geolocation of the Apple server is reflected as opposed to that of the device itself. However, if the ad network enriched the postback, geolocation is available. 

Site ID

The site ID as reported by the ad network

Additional filters

Click the predict_dashboard_plus_symbol.png symbol on the bar to add additional filter options including:

  • Adset
  • Agency
  • Source (organic/non-organic) – for AF model and SSOT attribution only

Bubble chart

The heart of the Predict dashboard is an interactive bubble chart that shows the distribution of predicted KPIs for new users. Many display options are available, giving you the ability to highlight the data most important to you. These options are described in detail in the table below.

Predict bubble chart components

        Item Description

A.png

 Data visualization options

predict_dashboard_KPI_dimension_view_options.png

Use the data visualization options to select:

  • the Predicted KPI to display
  • the Dimension by which you want the data broken down
  • the Characteristics by which you want the data grouped

Predicted KPI

By default, the chart displays bubbles based on pARPU value. You can select to display the bubble chart for any of the following KPIs:

  • pARPU
  • pROAS
  • p-% Retention D {x}, {y}, {z}
  • p-% Paying users

Dimension

Each row of the chart displays the bubbles broken down by the dimension you select.

Hover over the dimension description on the far left side of each row to see the total number of users in all bubbles on the row, including white bubbles (and gray bubbles, if applicable).

For example:

  • When the default view of Media source is selected, each row displays the distribution of bubbles for a single media source.
  • Hover over the name of each media source to see the total number of users attributed to that media source during the selected date range.

Available dimensions:

  • Media source (including a row for organic attribution when AF model or SSOT attribution is selected)
  • Adset
  • Agency
  • App ID
  • Campaign
  • Campaign ID
  • Channel
  • Geo (available for SKAN or AF model attribution)
  • Site ID
  • Source (available for AF model or SSOT attribution)

Characteristics

  • Each bubble represents a cohort that shares all the same characteristics.
  • All of the characteristics you select are applied to each bubble. Therefore, each characteristic you add to your selection generally results in more bubbles being displayed on the chart, with each bubble representing fewer users.
  • The selected dimension (see above) is always one of the selected characteristics.
  • Available characteristics:
    • Media source
    • Adset
    • Agency
    • Campaign
    • Campaign ID
    • Channel
    • Geo (available for SKAN or AF model attribution)
    • Site ID
    • Source (available for AF model or SSOT attribution)
  • Example:
    • The default selection of Media source and Campaign means that each bubble represents a group of users attributed to the same media source and campaign.
    • If you add the additional characteristic of Geo, each bubble now represents a group of users attributed to the same media source, campaign, and geo. Since there are fewer users that share all of these characteristics, each bubble represents fewer users, and more bubbles are required to capture all users.
B.png

Sorting options

predict_dashboard_sort_options.png

Use the sorting options to display the rows in the bubble chart (in ascending/descending order) by:

  • the total number of users in all bubbles on the row; or 
  • the row average for the selected KPI
C.png

Bubbles

Each bubble represents a group ("cohort") of newly-acquired users with the same characteristics (see A.png, above):

  • Hover over a bubble to display the cohort's characteristics, the number of users it includes, and the value of the selected KPI for these users.
  • The size of the bubble corresponds to the total number of users in the cohort, and the color of the bubble represents the predicted KPI value for these users.
  • Click on a bubble to open a side panel with additional details about the cohort, including a trend chart for the selected KPI, additional predicted KPI values, and the actual cost of acquiring these users (eCPI).
D.png Distribution scale Scale of predicted KPI values in the distribution
E.png Value filter Move the endpoints to narrow the range of predicted KPI values for which bubbles are displayed.
F.png Row average Average predicted KPI value for all bubbles on the row.
G.png Legend

Description of the chart's visual elements:

  • Cohort size: Each bubble on the chart represents a group of newly-acquired users with the same characteristics (see A.png, above). Bubble size corresponds to the total number of users in the cohort
  • Bubble color: Bubble color represents the predicted KPI value for the users in the cohort (higher predicted KPI value → darker bubble color)
    • When there is no prediction for more than 60% of users in the cohort, the bubble is colored white.
      • As long as there is a prediction for at least 1 user in the cohort, the white bubble is displayed within the regular distribution.
      • When there is no prediction for any of the users in the cohort, it is displayed as a no-prediction bubble in the no-prediction column of the chart, if you have turned on this option (see I.png and J.png, below).
    • When the selected KPI is pROAS and there is no cost data available for the cohort, a gray no-cost-data bubble is displayed in the no-prediction column of the chart, if you have turned on this option (see I.png and J.png, below).
H.png Minimum cohort size By default, the chart displays bubbles for all cohorts with 10 or more users. Use this option to select a minimum cohort size higher or lower than 10.
I.png No-prediction column selector

Select whether to show a column on the far right side of the chart that displays no-prediction and no-cost-data bubbles.

  • By default, this column is not displayed.
J.png

No-prediction bubbles

No-cost-data bubbles (pROAS only)

No-prediction bubbles and no-cost-data bubbles appear in the no-prediction column if you have turned this option on (see I.png, above).

  • A white no-prediction bubble is displayed when a cohort has no prediction for the selected KPI.
  • A gray no-cost-data bubble is displayed when the selected KPI is pROAS and there is no cost data available for the cohort.

Bubble details

predict_bubble_details_panel.png

Click on a bubble in the chart to open a side panel with additional details about the cohort.

  • Cohort details: the full description of the bubble based on the characteristics you have chosen (see A.png above).
  • Trend chart: a chart showing the predicted value of the selected KPI for cohort users who installed the app each day.
    • What does this mean? The chart breaks down the full cohort into groups of users who installed the app each day, and it displays the predicted KPI value for each of these date-based groups.
      • This allows you to consider the predicted KPI value for each day's users and how it has changed during the selected date range, making it particularly useful for analyzing the impact of campaign changes (such as different creatives, etc.)
      • Caution! Don't interpret the chart as showing the change in the predicted KPI value for the whole cohort over time. This is not what it is intended to represent.
    • Trend charts by metric:
      • pARPU and pROAS are displayed on the same trend chart (entitled pRevenue trend).
      • p-% Retention for all 3 retention measurement dates is displayed on the same trend chart (entitled p-% Retention trend).
      • p-% Paying users is displayed in its own trend chart (entitled, appropriately enough, p-% Paying users trend).
  • KPIs: predicted KPI values and eCPI for the cohort. Learn more about Predict KPIs and how they are calculated.
  • User count: the total number of users in the cohort including percentages for those:
    • with prediction
    • without prediction as a result of insufficient data
    • without prediction as a result of a postback that does not report a conversion value due to SKAN privacy thresholds (relevant for SKAN and SSOT attribution methods only)

Results table

Scroll down below the chart to view a customizable, downloadable results table, including all data fields for each cohort.

predictSK_v2_dashboard_results_table

        Item Description

A.png

 Grouping options
  • Use the grouping options to specify the characteristics used to group the data in the results table. Data must be grouped by at least one characteristic (the "primary characteristic").
  • The grouping options allow you to add, remove, and re-order the characteristics by which the data is grouped.
  • Each additional characteristic you select allows you to drill down further into the results data.
  • By default, the data is grouped by the following characteristics:
    • Media source
    • Campaign (within each media source)
B.png

Chart rows
with drill-down
  • The values of the primary characteristic are displayed as rows in the chart.
  • Click on the drill-down icon predict_dashboard_drill_down_icon.png to expand each row and display the values of each additional selected characteristic.
C.png

Column headers
  • Sort the table by any column by clicking its header.
  • Additional data fields (beyond those in the Bubble details panel) include pRevenue, p-% Whales and Total cost of acquiring the users in the cohort.
D.png

predict_dashboard_total_users_icon.png
  • Download the results table as a CSV file (for the date range and attribution method selected).
  • Each characteristic you select in the grouping options (see A.png above) will be displayed as a column in the downloaded file.

Validation report tab

Use the Validation report tab to compare the predicted results with the actual campaign results.

Based on the validation insights, you can see how accurate Predict is at giving you campaign forecasts, helping you determine if you should continue, stop, boost, or optimize your campaigns.

 

The following sections describe the main components of the Validation report tab.

Model information

The Model information shows when the model was last trained and the install date range of the users the model trained on. This is useful to know if changes or new features are added after the last trained version. The behavior of new users might be different than the users on the last trained model.

Headline metrics

KPI_Agg.jpg

Use the Headline metrics for a high-level view and comparison of the ARPU, paying users, and rolling retention information between the predicted and actual results.

Metric component layout

        Item Description

A.png

 Key metric name

KPI metric name

There are four KPI metrics available:

  • Total users
  • ARPU coverage 
  • Paying users coverage  
  • Rolling retention D7 coverage 
B.png

Coverage

The coverage ratio between the actual and predicted results.

This is calculated using the formula predicted/actual.

  • 100% means prediction matched actual results
  • If the percentage is less than 100%, the prediction was lower than the actual result
  • If the percentage is higher than 100%, the prediction was higher than the actual result

Example 1

The predicted result is $20.52 and the actual result is $23.24. The predicted result is lower than the actual result by $2.72 (12%), and the coverage is 88%. 

 

Example 2

The predicted result is $19.57 and the actual result is $15.37. The predicted result is higher than the actual result by $4.2 (27%) and the coverage is 127%.
C.png

Bullet graph

A visual bar showing the coverage. The dark blue line represents the predicted value and the vertical light blue line represents the actual value.

  • Left of the vertical light blue line, the prediction is lower than the actual result
  • Right of the vertical light blue line, the prediction is higher than the actual result
D.png

Numbers

 The predicted and actual results. Depending on the metric, this is displayed as a percentage or as a currency. 

 

Bubble chart

The bubble chart shows the distribution of predicted vs actual KPIs.

        Item Description

A.png

 Data visualization options

Select the Predicted vs actual KPI and then select the dimension to group it by. 

 

Predicted KPIs

By default, the chart displays bubbles based on ARPU value. You can display the bubble chart for the following KPIs:

  • ARPU
  • Paying users

Dimension

By default, the chart is grouped by Media source. You can group the bubble chart by the following dimensions:

  • Media source
  • Campaign
  • Geo
  •  
B.png

Bubbles

Each bubble represents a group ("cohort") of users.

  • Hover over a bubble to display the selected dimension, the number of users, and the values of the selected KPI for this user group
  • The size of the bubble corresponds to the total number of users in the cohort
C.png X and Y axis shows the predicted (y-axis) and the actual (x-axis) KPI values.

Results table

Validation_chart.jpg

        Item Description

A.png

 Grouping options
  • Use the grouping options to group the data in the results table.
  • Data can be grouped by the following:
    • Media source
    • Campaign
    • Geo
B.png

Chart rows
  • The values are displayed as rows in the chart.
C.png

Column headers
  • Sort the table by any column by clicking its header.
  • Additional data fields include pRevenue, Revenue, and Revenue coverage, seen in the Results table.
D.png

predict_dashboard_total_users_icon.png
  • Download the results table as a CSV file.

Notes

 

  • The bubble chart and Results table will only show data for cohorts with over 100 users.
  • The bubble chart and Results table are only available using the AF attribution model.

 

Putting it all together—a dashboard example

The following example demonstrates some of the capabilities of the Predict dashboard and how you can use these capabilities to efficiently analyze your data and take meaningful action.

 Example

Let's say you are a UA manager responsible for running campaigns in the UK, and you want to look at the predicted quality of users being driven by the campaigns you started last week. The main KPI you use to evaluate campaign performance is ROAS.

  • Using the filter bar, you set the date range to display the data for installs during the past 7 days, and you set the geo filter to display only data for the UK.
  • In addition to your new campaigns, you have many ongoing campaigns. For the moment, you want to focus only on the campaigns you started running last week, so you use the campaign filter to select only these campaigns.
  • You change the data visualization options to show the pROAS KPI. For now, you keep the other options set to the default values (Dimension = media source; Characteristics = campaign and media source)
    • This means each row represents a media source and each bubble on the row represents a cohort of users attributed to the same media source and campaign.
  • Even with the filters you've applied, there are lots of small bubbles that make it harder to analyze your data, so you change the minimum cohort size to 50. This allows you to focus only on the largest cohorts.
  • You notice that there are several bubbles on the left side of the distribution, meaning that they have low pROAS in comparison to other cohorts.
  • You decide that you want to concentrate now on only those cohorts with pROAS of less than 100%, so you slide the right endpoint of the value filter down to 100%.
  • You want to understand if there are specific site IDs that are pulling down the pROAS, so you change the characteristics selected in the data visualization options. You remove the campaign characteristic and add site ID instead.
    • After this change, each row still represents a media source, but now each bubble represents a cohort of users attributed to the same media source and site ID.
  • With the revised view, you quickly see that it's just a few site IDs that have low pROAS. When you click on these bubbles, the KPIs in the bubble details panel show that these cohorts actually have decent pARPU. The low pROAS is caused primarily by the high cost of acquiring these users (eCPI).
  • To further analyze all the data, you scroll down to the results table set the grouping options to show media source > campaign > site ID. You download the table as a CSV file, to which you can apply your own pivot tables and BI analysis.

Based on the insights you obtain from this analysis, you might take one or more of the following actions to optimize your new campaigns:

  • Stop running the campaigns on low-performing media sources or lower the bids. (With lower bids, it might make sense to continue running the campaigns on these media sources.)
  • Allocate budget to more productive media sources.
  • Adjust campaign configuration on the low-performing media sources (for example, by targeting better-performing site IDs) to balance the quality of the users they provide with the cost of acquiring them.