Frequently Asked Questions about the Measurement suite of analytic reports.

What is the Measurement product?

The Measurement product is a new set of analytic reports available within the TAdS MAP platform that combine the proprietary T-Mobile app engagement (mobility) data with selected ad log data to create metrics that focus on mobile app lift.

Measurement is comprised of five (5) core analytic reports:

• User Acquisition - Mobile App Installation Lift
• Exposed Engagement - Mobile App Usage Lift
• Exposed Growth - Mobile App Adoption Lift
• Target Acquisition - Audience Segment Installation Performance.
• Reach + Frequency - KPI Performance by Frequency Bucket

I don’t see a the Advertiser or Campaign I am interested in?

The TAdS Client Success/Enablement team can most likely add missing Advertisers and Campaigns to Measurement for you. The Measurement product has the ability to report on any campaign that was...

• ... activated on a participating DSP.
• ... active on/after July 1, 2021.
• ... meets the minimum device panel count to be reportable.

Assuming that the above 3 qualifications are met, then the following information will need to be provided to your Client Success/Enablement representative:

• Provide the name of the Advertiser you would like reporting for.
• Provide the campaigns you would like reporting for, as well as any rollups of those campaigns.
• Provide the mobile apps that you would like to be reportable against these campaigns.

🚧

New Data Availability

New Campaign data processing for the Measurement reports occurs on the Friday-Sunday of every week. New campaigns and mobile apps added between Monday-Friday of a given work week, will be available with Measurement report data by the following Monday.

More campaigns, campaign rollups, and mobile apps can be added at any time for processing, but will follow the above data creation schedule.

I believe that there is missing data, can it be fixed?

Missing data can often be resolved. If you believe that not all active dates are represented with data for a given campaign, you can report the issue via your Client Success/Enablement representative.

Does Exposed Growth + Un-Exposed Growth = Overall Growth?

The 3 different flavors of growth (Exposed / Unexposed / Overall) are all calculated based on their own universes, and therefore the percentages will not total to each other. For example, suppose that "Streaming Service A" is running a campaign with TAdS in January 2023. Let’s also suppose that…

On January 10th that the number of ad-exposed "Streaming Service A" owning mobile devices was 100,000, while the number of unexposed "Streaming Service A" owning devices was 900,000. Its safe to assume that 1 million mobile devices own the "Streaming Service A" app on this date:

• (900,000 + 1000,000 = 1,000,000)

On January 11th, the number of ad-exposed "Streaming Service A" owning mobile devices was 150,000, while the number of unexposed "Streaming Service A" owning devices was 800,000. Its safe to assume that 950,000 mobile devices own the "Streaming Service A" app in this scenario:

• (900,000 + 1000,000 = 1,000,000)

The mobile growth from January 10th to January 11th for the "Streaming Service A" mobile app would then be:

• Exposed: (150,000 - 100,000) / 150,000 = + 33.33% Growth
• Unexposed: (800,000 - 900,000) / 800,000 = - 12.5% Growth
• Overall: (950,000 - 1,000,000) / 950,000 = - 5.26% Growth

From this example we can see that changes in ownership can create different growth percentages that do not sum together, while the counts themselves still do add up.

How is the data created, and does it align to Insights data?

The Measurement report data is created through a series of scaling steps, similar to the way in which App Insights data is created. The scaling steps are in place to assure that T-Mobile’s Android only device data is representative of iOS and other mobile carriers (full US footprint of mobile devices).

It will always be the case that ad-exposed devices performing a KPI action (like installing a mobile app) in Measurement are always less than the Insights equivalent metric value, because not all installing devices can be exposed during a campaign. The Measurement data is aligned to the Insights data in this way.

More In Depth Explanation

All devices served an impression from a given campaign are collected into a device pool. This ad-exposed device pool is then intersected with the raw Insights mobile app engagement reporting device panel. Those devices which overlap between the ad-exposed device pool and the Insights mobile app engagement panel become the Measurement Report Panel. The Measurement Report Panel is then scaled up in such a way that it will be a representative portion of the scaled Insights data.

For example

Assume that within the Insights mobile app engagement data, there are 100,000 installations of the "Streaming Service A" mobile app.

Assume also, that of those 100,000 install events that were captured, 1,000 of those devices were found to also have been exposed to an ad from the "Streaming Service A" campaign via a device level match between ad log and the mobile app engagement panel.

Within Insights, the initial 100,000 mobile app installation events for "Streaming Service A" will be scaled up to represent the full mobile universe, let’s say that these 100,000 devices scale up to show a count of 1,000,000 mobile devices in Insights.

Similarly to how the mobile app installation events for "Streaming Service A" are scaled up to be more indicative of the total US mobile device owning population for Insights, so will the 1,000 matched ad-served devices be scaled up for Measurement. These 1,000 devices will be scaled up in such a way as to be the correct portion of the 1 million devices reported in Insights. So the final Exposed Install count for Measurement will be, for example, 10,000.

Within these matching and scaling processes precautions are taken to ensure that the data is always telling the correct story:

• Measurement Ad-Exposed Installing devices must be proportionally less than the total number of Installing devices in Insights.
• The Ad-Exposed Installing devices can never be more than the reach of the campaign (number of distinct devices served an impression).
• The Incremental Installing Devices can never be more than the Exposed Installing devices.

The above are merely a small example of the data checks that are performed during the Measurement report creation process.

Is there reporting for TAdS O&O products and Rideshare?

The short answer is: Not yet, but hopefully soon.

The initial core five (5) Measurement reports are all in service of Magenta Audience Network (mAN) programmatic campaigns. However, work is underway to develop and provide analytic campaign lift and full funnel performance reporting for Native App Install campaigns as well.

How do I gain access to Measurement?

Measurement access can only be granted by your Client Success/Enablement representative. Please contact this representative for more information on how to gain access to Measurement reporting.

Is Measurement reporting data validated?

To-date, only the Insights data has undertaken benchmarking tests against competitors in the mobile app reporting space.

Can I select other attribution Impression Windows?

For the time being, Measurement reports only allow reporting with either a 30 (preset), 60 or 90 day Impression (attribution) Window. In order to guarantee as robust a report as possible, there is a minimum 30 day lookback window in place to maximize the strength of device intersection between the mobile app engagement panel and the ad-exposed devices in a given campaign log.