Steps for Accurate Cross-Device, Cross-Channel Identity Matching

By Published on .

Accurately analyzing user identities across channels and devices is a critical capability -- and difficult.
Accurately analyzing user identities across channels and devices is a critical capability -- and difficult. Credit: anyaberkut/iStock

As with many measurement technologies that emerge to service rapidly evolving media platforms, technologies that enable marketers to match consumer identities across devices and multiple platforms are characterized by an array of unclear and obfuscated approaches.

The reality of these varying methodologies is that marketers, in order to succeed, need to understand the differing approaches and know what the right questions to ask vendors.

To address this need, The Coalition for Innovative Media Measurement, or CIMM, recently set out to define the state of the art in census-based methods for establishing consumer identity across multiple platforms and Internet-connected digital devices. Our research, which was based on interviews with advertisers, agencies, media companies and research vendors, found that for the foreseeable future, marketers will have to work with a variety of solutions. Specifically, they will have to deploy deterministic, probabilistic and increasingly hybrid approaches in order to create best-in-class solutions that combine scale with accuracy.

A full copy of this research can be downloaded here, but the following are the critical considerations marketers need to keep in mind as they start working with cross-device and cross-channel identity matching vendors:

  • Understand that access to "truth files," documents known to be 100% accurate, and data points on which solution providers match are the key variables influencing match rates.
  • Have an awareness that key measures to understanding the true implication of match rates are accuracy and precision, followed by reach and scale.
  • Understand the individual data points that contribute to overall match rates, and their respective strengths and weaknesses.
  • Have a firm grasp on how providers weigh and balance data in their various graphs, and know how to measure the ability of those graphs to target the correct consumers across devices and platforms.
  • Appreciate that the intersection of increased government scrutiny and heightened consumer concerns around privacy means marketers using identity-based marketing should have a deep understanding of the data sources they are working with, and how they do or do not protect consumer privacy.

This overlapping and confusing ad tech ecosystem needs further industry focus and development if it is to better serve the needs of marketers. In light of that, CIMM is calling for the following:

  • Third-party validation of key measures such as match rates and accuracy.
  • Industry-level education in terms of best practices for identity matching and assessment of data quality.
  • Stronger clarity and standards about protecting consumer privacy.
  • Industry standardization definition of key measures, and agreement on how they should be calculated.
  • Industry-level education in terms of best practices for identity matching,

The ability to measure and accurately analyze user identities across channels and devices is a critical capability. Without it, the industry cannot hope to keep up with consumers, who are changing viewing habits as quickly as platforms allow.

The important contribution marketers can make now is to ensure that they understand fully the methodologies being used, the questions to ask -- to guarantee accurate matches are being made -- and how to push vendors to further refine technologies that both enhance scale and accuracy, while properly protecting the privacy of consumers.

In this article: