Incrementality guide

AppsFlyer Incrementality employs 2 alternative methods for calculating lift:

• Intent-to-treat
• Reach-based

While reach-based methodology is preferable for purposes of evaluating campaign performance, its calculation requires the measurement of data points not provided by all ad networks. For these networks, intent-to-treat methodology is used instead.

ITT-based lift

Intent-to-treat (ITT) experimental methodology has its foundations in medical science. Using ITT, researchers randomly split a given population between a control group who are not treated and a test group whom researchers intend to treat. Researchers measure the efficacy of a given treatment by comparing the results of the groups. Whether or not a test group member actually received treatment is not taken into account.

Similarly, in remarketing, the lift metric measures campaign efficacy. It doesn't matter if a user in the test group actually engaged with a campaign (received treatment) because it is the intent that is relevant in calculating incrementality. In other words, results are calculated as if every member of the test group actually engaged with the campaign.

When is it used?

AppsFlyer Incrementality uses ITT-based lift calculation for SRNs.

Although it is not the preferred method for measuring lift in our context, ITT remains a scientifically sound experimental method, and it provides valid results when reach-based calculation is not possible.

Interpreting results

The underlying assumption that all members of the test group were actually reached by a campaign often results in low conversion rates and, as a consequence, low incremental lift.

Reach-based lift

As its name implies, reach-based lift calculation takes into account whether or not a member of the test group was actually reached (served a campaign ad) prior to converting. In addition to requiring additional data, reach-based lift calculation is also more complex in that it requires certain mathematical adjustments to the control group in order to be statistically valid. These adjustments are cumulatively known as the creation of a "counterfactual" control group.

When is it used?

AppsFlyer Incrementality uses reach-based lift calculation for non-SRNs (attribution-link networks).

While it is the preferred method for calculating lift, reach-based calculation requires that ad networks report the following data to AppsFlyer:

• Impression data
• Campaign ID (from the attribution link)

If this data is not reported, incrementality measurement is not available (and displays as N/A on the Incrementality dashboard).

Interpreting results

Since reach-based lift calculation takes into account only test group members that were actually reached by a campaign, networks that are measured using reach-based incrementality will usually show amplified results (positive lift will become much higher, and negative lift will become much lower).

Best practices

Because the 2 different lift calculation methodologies can yield significantly different results, the following best practices are recommended in order to make interpretation more meaningful:

• When creating your experiments, select audiences that are split only among a control group and one or more networks of the same type, either:
  • Attribution-link networks (non-SRNs) for reach-based calculation; or
  • SRNs for ITT-based calculation
• In your initial experiments, consider using an audience split only between a single network and a control group. This makes it less likely that comparisons will be made among 2 or more networks for which different lift calculations are used.

When evaluating the validity of Incrementality experiments, sound scientific methodology requires that we consider statistical significance (the level of confidence that the lift result isn't due to random events or chance). 

On the Incrementality dashboard, significance is reported as a percentage for each media source:

• The percentage indicates the likelihood that the results for that media source would repeat themselves if the experiment were performed again under similar circumstances.
• If statistical significance is determined to be below 60%, the dashboard will display Unreliable results instead of a percentage.
• Significance is affected by audience size, experiment duration, and media source performance.

For test-group-level metrics (as opposed to per-user metrics), lift is calculated based on a scaled control group. In other words, the size of the control group is adjusted to match the control group size for each media source.

To access the Incrementality dashboard:

1. In AppsFlyer, go to Dashboards > Incrementality.
2. From the experiments list, click the row of the experiment for which you want to see the dashboard.

	Component	Description
	Filter bar	Measure: Select the metric you want to view in the dashboard: Conversions – The number of unique users who performed the selected event Revenue – Revenue generated by converting users of the selected event Events – The number of occurrences of the selected event performed by converting users Your selection in the Measure filter determines which metrics are displayed throughout the dashboard.   Event: Select the in-app event for which you want to view data in the dashboard.   Media source: Filter the dashboard display by one or more media sources tested in the experiment.   Date range: Filter the incremental analysis of the audience according to the selected date range.
	Significance tile	Displays the statistical significance of the results shown in the dashboard
	Headline metric tile	The metric displayed is controlled by your selection in the Measure filter, as shown below. Measure filter selection Displayed headline metric Conversions/Events Cost per incremental action (CPiA) Average cost of every incremental event, based on campaign cost Revenue Incremental return on ad spend (iROAS) Ratio of incremental revenue generated for every dollar spent on the campaign Note: Calculation of metrics displayed in the headline tile require cost data. Therefore, these metrics are available only to Xpend subscribers. If the required cost and/or revenue data is not available for this or any other reason, the relevant metric will display as N/A.
	Target size and reach tile	Target size: The number of distinct users, for each media source, who were added to the audience for the first time during the experiment's date range Also represented by the pie chart's inner ring (darker color) Reach: The percentage of the target that actually saw an ad from the relevant media source's campaign Also represented by the pie chart's outer ring (lighter color)
	Test-group-level lift tile	Displays total incremental metric and lift for the test group (by media source): Total incremental lift is calculated using a scaled control group. The metric displayed is controlled by your selection in the Measure filter, as shown below. Measure filter selection Displayed metric Conversions Total incremental conversions Events Total incremental events Revenue Total incremental revenue
	Per-user lift tile	Displays incremental metric and lift per user (by media source): Per-user incremental lift is calculated using the actual number of users in the test group and control group. The metric displayed is controlled by your selection in the Measure filter, as shown below. Measure filter selection Displayed metric Conversions Incremental conversion rate Events Incremental events per user Revenue Incremental revenue per user
	Trend chart	Displays the selected per-user incremental metric/lift over the date range of the experiment: The displayed metric is controlled by your selection in the Measure filter and is the same metric as the one shown in the Per-user lift tile. A dotted line (in the chart view) indicates partial data for the reported date (meaning future events can still affect the data). Select various view and download options using the Trend chart view controls (as further described immediately below).
	Trend chart view controls	Use the view controls to select options for Trend chart display and download: Cumulative or daily (On-day) values Chart or table format   Use the Export button to download the data as a table in CSV format. The data downloaded is controlled by your selection with the view controls: Cumulative or On-day

Category	Report name	Data Locker folder
First-seen	First-seen users	incrementality_first_seen_users
In-app events	Organic in-app events	incrementality_organic_inapps
	In-app events non-organic	incrementality_inapps
	In-app events re-attributions	incrementality_inapps_reattr
Sessions	Organic sessions	incrementality_organic_sessions
	Sessions non-organic	incrementality_sessions
	Sessions re-attributions	incrementality_sessions_reattr
Uninstalls	Uninstalls (not currently populated)	incrementality_uninstalls

The basic logic underlying report structure is as follows:

• Users participating in the experiment—First-seen report
  • In Audiences, rules are set characterizing users to include in the experiment.
  • When a given user, is identified as matching the rules, the event is recorded in the first_seen report. 
    • Users are allocated randomly to a test or control group indicated by the is_control_group field.
    • Users in the test group are allocated to a media source (pid_destination) for retargeting.
• User engagement within the app: User engagement with the app during the experiment is recorded in context-specific reports: 
  • Engagement type: Session or in-app event
  • User attribution status when the user is first-seen: Organic, non-organic, re-attribution. For example, in the past, a user installed the app and was attributed to organic. As such, during an experiment, the attribution status is organic.  
• Uninstalls: Users uninstalling the app during the experiment. Uninstall measurement must be active. 

Field availability varies according to report type as indicated.

Fields unique to Incrementality (per report type)

Field	Description	First seen	In-app events	Sessions	Uninstalls
is_control_group	If true, the user is part of the control group	Y	Y	Y	Y
pid_destination	The media source the user is sent to	Y	Y	Y	Y
audience_id	Unique identifier	Y	Y	Y	Y
joined_audience_date	Date user first joined the audience	Y	Y	Y	Y
audience_name	Audience name (not unique)	Y	Y	Y	Y
tm	Hour of day		Y	Y	Y
timestamp	Event time stamp YYYY-MM-DD HH:MM		Y	Y
app_ids	App ids associated with the audience rules	Y

 

Other Incrementality report fields (per report type)

Field	Display name*	First seen	In-app events	Sessions	Uninstalls
advertising_id	Advertising ID (GAID)	Y	Y	Y	Y
android_id	Android ID		Y	Y
app_id	App ID	Y	Y	Y	Y
app_name	App name		Y	Y
app_version	App version		Y	Y
appsflyer_id	Appsflyer ID		Y	Y	Y
revenue_alt	App-specific currency		Y
bundle_id	Bundle ID		Y	Y
country	Country code		Y	Y	Y
currency	Currency code		Y	Y	Y
customer_user_id	Customer user ID		Y	Y
brand	Device brand		Y	Y
device_category	Device category		Y	Y
model	Device model		Y	Y
device_model	Device model		Y	Y
device_type	Device type		Y	Y
event_name	Event name		Y	Y	Y
event_revenue	Event revenue		Y
event_revenue_currency	Event revenue currency		Y
event_revenue_u_s_d	Event revenue USD		Y
event_time	Event time		Y	Y	Y
event_value	Event value		Y	Y
idfa	IDFA	Y	Y	Y	Y
idfv	IDFV		Y	Y
imei	IMEI		Y	Y
is_purchase_validated	Is receipt validated		Y
os_version	OS version		Y	Y
platform	Platform		Y	Y	Y
sdk_version	SDK version		Y	Y
* According to raw data specification