Measurement - Glosario

A glossary of frequently used terms and phrases in the Measurement reporting ecosystem.

An ad Impression is an ad spot in an ad campaign that has been served to a device on any given ad channel by the activating DSP during a specific KPI Date Range. Ad Impression metrics can be found at both the Snapshot (aggregate) and Info Table / Performance Analysis (detail) levels across most Measurement reports, including...

• Measurement Dashboard
• User Acquisition
• Exposed Engagement
• Reach + Frequency

When ad Impressions are summed they are reported as Gross Impressions. In a report Snapshot the Impressions are summed from the daily level. When ad Impressions are averaged across time periods they are reported as [Daily / Weekly / Monthly] Avg Impressions.

Percent Impressions is a metric found in the Reach + Frequency report. This metric provides the proportion of total impressions served for a campaign by Frequency bucket.

In a graph in which the X-Axis is the number of impressions served to a device, Frequency, and the Percent Impressions at Frequency bucket 3 is 50%, this means that 50% of all impressions served in the campaign went to devices that received between 1 and 3 ads from the campaign.

The Reach of a campaign is the number of distinct (unique) devices that were served ads (exposed) during the KPI Date Range + Impression Window.

When the Reach is presented in a report Snapshot, it is always presented as the Average Daily Reach. The Average Daily Reach is the sum of all daily "Reach in Window" values over the number of days in the KPI Date Range + Impression Window.

Próximamente...

Percent Reach is a metric found in the Reach + Frequency report. This metric provides the proportion of distinct (unique) devices reached by a campaign by Frequency bucket.

In a graph in which the X-Axis is the number of impressions served to a device, Frequency, and the Percent Reach at Frequency bucket 3 is 50%, this means that 50% of all devices reached by the campaign received between 1 and 3 ads from the campaign.

Frequency is the number of times a device is exposed to ads from a given campaign during the time that the campaign is active. Frequency is presented in Measurement through the Reach + Frequency report, in which devices are grouped together based on the number of impressions they received from a campaign.

The ad Spend value is the number of US dollars spent by the advertiser to serve ad impressions to devices for the selected campaign and KPI Date Range.

When Spend is calculated in a report Snapshot, all daily Spend values are summed together.

A Click is an end user action, in which the device user has engaged an ad spot they have been served by touching or clicking on it. Click events are recorded across all ad channel by the activating DSP for the selected campaign and date range. Note that Click counts are not differentiated between goal and non-goal clicks.

CPC (Cost per Click) is calculated by taking the total Spend over the sum number of Clicks:

• (Total Spend / Sum Clicks) = CPC

CTR (Click Through Rate) percentage is calculated by taking the sum of Clicks over the Gross Impressions, and multiplying the resulting value by 100:

• (Sum Clicks / Gross Impressions) x 100 = CTR %

An Exposed Device is a device that is served an ad impression from a campaign, regardless of the ad channel by which that impression is served.

A device is considered exposed to the ad if the ad was delivered within the selected KPI Date Range + Impression Window (30, 60 or 90 days lookback).

A device is considered to be a mobile app installer if that device downloaded the mobile app during the selected KPI Date Range of the Measurement report.

In Measurement reports, the installing devices are reported as either All Installs or Exposed Installs. The "All Installs" metric is the sum of all mobile app installs during the KPI Date Range from among all devices in the US mobile universe. This value matches the Insights > Installation > Total Installs value.

The "Exposed Installs" metric is the sum of all mobile app installs during the KPI Date Range from among all devices that were served an ad impression from the selected campaign during the KPI Date Range + Impression Window.

Percent Exposed Installs is a metric found in the Reach + Frequency report. This metric provides the proportion of Exposed Installs of a campaign by Frequency bucket.

In a graph in which the X-Axis is the number of impressions served to a device, Frequency, and the Percent Exposed Installs at Frequency bucket 3 is 50%, this means that 50% of all Exposed Installs associated to the campaign came from devices that received between 1 and 3 ads from the campaign.

The Installation Conversion Rate is calculated by taking the rate of mobile app installation among exposed devices (Exposed Install Rate) against the rate of mobile app installation among all US devices that are both exposed and unexposed (Overall Install Rate).

These rates help to show how often the mobile app was installed among both device groups:

• (Exposed Install Rate / Overall Install Rate) x 100 = Installation Conversion Rate (aka Exposed Installation Index)

The larger the value of the Installation Conversion Rate, the higher the rate that ad-exposed devices (those reached) installed the mobile app, as compared to installs from all US mobile devices in general.

These rates help to show how often the mobile app was installed among both device groups:

• (Exposed Install Rate / Overall Install Rate) x 100 = Installation Conversion Rate (aka Exposed Installation Index)

The larger the value of the Installation Conversion Rate, the higher the rate that ad-exposed devices (those reached) installed the mobile app, as compared to installs from all US mobile devices in general.

An Installation Conversion Rate value of “1.33x” can be read as: Those people exposed to an ad from this campaign installed this mobile app at a rate 1.33 times (33%) greater than the rate of installs from among the US mobile device populace in general.

Estimated Incremental Installs are calculated by taking the Installation Conversion Rate against the number of Exposed Installs:

• (Exposed Installs) x (1 - (1 / Installation Conversion Rate)) = Estimated Incremental Installs

Estimated Incremental Installs can be best read as the number of mobile app installs that occurred among ad exposed devices during the KPI Date Range, which are believed to be driven by exposure to the ad.

With an Installation Conversion Rate of 1.33x, and a count of Exposed installs of 14,000, the Estimated Incremental Installs will be 3,474.

Another way to read the Installation Conversion Rate value of 1.33x is as a ratio. An value of 1.33 is equivalent to a ratio of 4 to 3 (4:3). Meaning, that for every 4 Exposed Installs of a mobile app that occurred, 1 of those 4 was determined to be an Estimated Incremental Install that was more likely to be driven to app adoption after having been exposed to an ad from the campaign.

We can also speak about the Installation Conversion Rate value of 1.33x as a percentage of Estimated Incremental Installs to Exposed Installs. To speak about the data this way, we will utilize a portion of the above Estimated Incremental Installs calculation:

• 1 - (1 / Installation Conversion Rate) = 1 - (1 / 1.33) = 1 - (.75) = 1 - .75 = .25 = 25%

In the above case, 25% of Exposed Installs were Estimated Incremental Installs. Similarly, if you had an Exposed Install Index of 3, that would mean that 66% of Exposed Installs were Estimated Incremental Installs.

A mobile device is "active" within a given app if that app is...

• ...'in focus' on a device...
• ...the only app that can been seen and interacted with on the device's screen at that point in time.

All engagement data in Measurement is tied to the existence of a Focus Event record for a given app on a device. This includes foundational measurements such as app opens and app dwell time.

Measurement additionally utilizes a minimum dwell time threshold on all Focus Events, in which any Focus Event that is less than 2 continuous seconds (< 2 seconds) in duration is dropped from all data aggregations. This ensures that Measurement is only presenting intentional focus events in its reports, removing those events that may be accidental or otherwise unintentional.

Active Devices are sometimes also known as "DAUs (Daily Active Users)" or "MAUs (Monthly Active Users)", and is a 'reach' style metric that is based on app engagement. Active Devices is the count of all unique/distinct mobile devices that engage a given app in the date grain period, and during the KPI Date Range, being considered. Mobile devices qualify as engaged if they have at least one (1) focus event of two (2) seconds or longer in duration in the app.

The Exposed Active Devices metric is the count of all devices engaging with a mobile app, during the KPI Date Range, after having been exposed to an ad from the selected campaign during the KPI Date Range + Impression Window.

Percent Exposed Active Devices is a metric found in the Reach + Frequency report. This metric provides the proportion of Exposed Active Devices of a campaign by Frequency bucket.

In a graph in which the X-Axis is the number of impressions served to a device, Frequency, and the Percent Active Devices at Frequency bucket 3 is 50%, this means that 50% of all Exposed Active Devices associated to the campaign came from devices that received between 1 and 3 ads from the campaign.

The Active Device Conversion Rate is calculated by taking the rate of mobile app engagement among exposed devices (Exposed Active Device Rate) against the rate of mobile app engagement among all US devices that are both exposed and unexposed (Overall Active Device Rate).

These rates help to show how often the mobile app was engaged with among both device groups:

• (Exposed Active Device Rate / Overall Active Device Rate) x 100 = Active Device Conversion Rate (aka Exposed Active Device Index)

The larger the value of the Active Device Conversion Rate, the higher the rate that ad-exposed devices (those reached) engaged with the mobile app, as compared to engagement from all US mobile devices in general.

An Active Device Conversion Rate value of “1.33x” can be read as: Those people exposed to an ad from this campaign engaged this mobile app at a rate 1.33 times (33%) greater than the rate of engagement from among the US mobile device populace in general.

Estimated Incremental Active Devices are calculated by taking the Active Device Conversion Rate against the number of Exposed Active Devices:

• (Exposed Active Devices) x (1 - (1 / Active Device Conversion Rate)) = Estimated Incremental Active Devices

Estimated Incremental Active Devices can be best read as the number of ad-exposed devices engaging with a mobile app during the KPI Date Range, which are believed to be driven by exposure to the ad.

With an Active Device Conversion Rate of 1.33x, and a count of Exposed Active Devices of 14,000, the Estimated Incremental Active Devices will be 3,474.

Another way to read the Active Device Conversion Rate value of 1.33x is as a ratio. A value of 1.33 is equivalent to a ratio of 4 to 3 (4:3). Meaning, that for every 4 Exposed Active Devices on a mobile app, 1 of those 4 was determined to be an Estimated Incremental Active Device that was more likely to be driven to app engagement after having been exposed to an ad from the campaign.

We can also speak about the Active Device Conversion Rate value of 1.33x as a percentage of Estimated Incremental Active Devices to Exposed Active Devices. To speak about the data this way, we will utilize a portion of the above Estimated Incremental Active Device calculation:

• 1 - (1 / Active Device Conversion Rate) = 1 - (1 / 1.33) = 1 - (.75) = 1 - .75 = .25 = 25%

In the above case, 25% of Exposed Active Devices were Estimated Incremental Active Devices. Similarly, if you had an Exposed Active Device Index of 3, that would mean that 66% of Exposed Active Devices were Estimated Incremental Active Devices.

The Re-Engagement Rate is the percentage of those mobile app owners that were both inactive (dormant) in the app for at least 30, 60, or 90 days prior to the selected KPI date range, and whom then re-engaged with the mobile app during the KPI Date Range after exposure to an ad from the selected campaign.

When presented in Report Snapshot widgets, the value is the average of all daily, weekly, or monthly Re-Engagement Rates within the selected KPI Date Range.

The Overall Dwell Time Minutes metric is the sum of all minutes that a mobile app was in focus, across all devices focusing on the mobile app during the day, week, or month being considered. Dwell Time Minutes are associated to focus events, and Dwell Time Minutes are only counted towards the total if the focus event they are associated with is two (2) seconds or longer in duration.

Overall Average Dwell Time Minutes is the average minutes spent engaging with a mobile app per device. This value matches the Insights > Engagement > Avg Dwell Time (min). The calculation is:

• (Total Dwell Minutes) ​/ (Unique Engaged Devices)​​

The Exposed Dwell Time Minutes metric is the sum of all minutes spent in a mobile app during the KPI Date Range from among those devices that were exposed to an ad from the campaign during the KPI Date Range + Impression Window. Dwell Time Minutes are associated to focus events, and Dwell Time Minutes are only counted if the focus event they are associated with is two (2) seconds or longer in duration.

The Exposed Average Dwell Time Minutes metric details the average number of minutes an ad-exposed device spent in the mobile app per date period. The calculation is:

• (Exposed Dwell Time Minutes) ​/ (Reached Devices)

The Snapshot sub-value of the Exposed Average Dwell Time Minutes metric is the percent difference between the Overall Avg Dwell Time Minutes and the Exposed Avg Dwell Time Minutes. A positive number indicates that those exposed to an ad had a higher average in-app dwell time than all US mobile devices in general.

Overall Growth is the change (% Difference) in mobile app ownership from date period to date period from among all devices in the US mobile universe, both exposed and unexposed. This value matches the Insights > Growth > % Growth value. The calculation is:

(Period 2 Mobile App Ownership - Period 1 Mobile App Ownership) / Period 2 Mobile App Ownership) x 100

Within the Snapshot, all periodic Growth values are averaged together to arrive at a single overall value for the KPI Date Range.

The calculation for App Growth can result in negative growth when appropriate. Negative growth occurs when the number of Uninstalls meets or exceeds the number of Installs, resulting in app ownership changes that are stagnant or declining.

Exposed Growth is the change (% Difference) in mobile app ownership from date period to date period from among all ad-exposed devices (Campaign Reach). Within the Snapshot, all periodic Exposed Growth values are averaged together to arrive at a single Exposed Growth value for the KPI Date Range.

The Snapshot sub-value is the percent difference between the average Overall Growth and the average Exposed Growth. A positive number indicates that ad-exposed mobile devices had a higher average app adoption rate than devices in the US Mobile Universe in general.

Unexposed Growth is the change (% Difference) in mobile app ownership from date period to date period from among all non ad-exposed devices. Within the Snapshot, all periodic Unexposed Growth values are averaged together to arrive at a single Unexposed Growth value for the KPI Date Range.

The Response Rate metric is found in the Target Acquisition report. The Response Rate is the calculated as:

• (Exposed Install Rate / Unexposed Install Rate)

These installation rates are similar to those found in the User Acquisition report, which contribute to the creation of the Installation Conversion Rate in that report.

While the installation rates in Target Acquisition are similar in nature to those found in the User Acquisition report, the Target Acquisition installation rates are based on those devices in a specific targeted audience segment that were exposed to an ad from the campaign, and those that were not exposed to an ad from the campaign.

In this way, the unexposed installation rate operates as a holdout set of devices from the original audience segment, but which were not exposed to an ad from the campaign. While the exposed install rate operates as the target group that was exposed. When taken against each other, these two rates form a final Response Rate metric, which shows how the ad-exposed portion of the target performed against the KPI when compared to the non ad-exposed portion of the target.