Measurement: Closing the Identity Loop

As the digital ecosystem shifts to privacy-first technologies and approaches, measurement tactics that rely on identifiers will evolve or disappear. 

Many marketers worry about what that will mean for the precision of their measurement and attribution — and, no doubt, the marketplace is shifting. But where are we headed? 

Sales lift, which may be tied to an identifier today, will still be available in the future via modeled or panel-based techniques. Similarly, footfall measurement, which currently relies on location data, will likely be transformed. 

In the future, marketers will measure footfall through modeled conversion which could yield probabilistic results, or through a panel-based study that would provide the marketer with explicit findings. And while marketers may not be able to target users at specific locations, they may be able to drive users to specific locations and measure performance.

In the coming years, we see measurement of brand and performance use cases evolving into four main categories: 

• Panel-based measurement
• Indirect feedback loop
• On-device measurement
• Modeled conversions

Panel-based measurement (brand use cases): 

Brands and publishers can lean on their relationships with customers by turning their user bases into panels that they can solicit for direct feedback about products, services, and content. For example, YouTube runs panel-based ads that create a continuous feedback loop within its TV products. Brands and publishers can also use surveys to ask users if they are familiar with a specific brand or how likely they are to purchase products from that brand.  

Indirect feedback loop (brand-performance use cases): 

Indirect feedback consists of signals or data points about customer interactions that can be collected and used to measure the effectiveness of ad spend or inventory yield. For example, indirect signals could include information about how users respond to ads, how much time they spend on a page, or even aggregated conversion information.

On-device measurement (brand-performance use cases): 

On-device measurement matches conversions to user interactions with app ads. On-device measurement capabilities can create audiences at the device level and make the audience segments, not the individuals, available for targeting. As an example, Apple’s privacy-focused attribution tool SKAdNetwork helps ad networks attribute app installs at an aggregated level without using the Identifier for Advertising (IDFA). When ads are displayed for three seconds, the app notifies SKAdNetwork, which documents a successful view. If there is any engagement with the ad, the advertised app StoreKit is rendered, which is recorded by SKAdNetwork. If the app is installed during the SKAdNetwork attribution window, the ad network receives credit for the install, and the device sends the install postback to the ad network and a copy to the advertiser.

Modeled conversions (performance use cases): 

Methodologies such as media-mix modeling can provide marketers with a holistic view. The IAB defines marketing mix modeling (MMM) as a statistical analysis of aggregate sales, marketing, and business drivers data that quantifies the impact of different marketing channels and tactics (the marketing mix) on financial outcomes over time. The result is insights and recommendations that can be used to optimize marketing investment allocations and predict future outcomes. Self-attributing networks, where an ad network or platform like Google or Facebook models conversions that it can take credit for, may also fit within this category. 

The measurement and attribution landscape is evolving alongside the identity landscape, and it’s important for marketers to stay abreast of the strategies and tactics that will remain viable in the future. 

For a detailed look at how the identity landscape is shifting, and what that means for advertisers and publishers alike, download IDENTITY: DECODED, a comprehensive guide to identity in ad tech.