CIMM Probes Differences in TV Attribution Models

Research-on-research body the Coalition for Innovative Media Measurement (CIMM) has announced it is working on a TV Attribution study, aiming to define best practices for better representation of television in attribution models.

The CIMM is comprised of leading video content providers, media agencies and advertisers, and its research explores two main areas: the current and future potential of television measurement through the use of return-path data, and new methods for cross-platform media measurement. For the new study, it is working with brand and media metrics consulting firm Sequent Partners and TV and streaming digital specialist Janus Strategy & Insights.

The study will help define best practices for representation of television in attribution models. Attribution providers use different modeling techniques and data sources, using both STB data and Smart TV ACR data - leading to different results and business decisions. The partners will look at six national, linear television campaigns that aired in 2019, to assess what drives the difference in attribution results, and will also be assisted by companies including Alphonso, Ampersand, Comscore, Hive, iSpot, Kantar, Nielsen, Samba and TVSquared. CIMM says learnings will include what contributes to differences in the data that detect TV ad occurrences, how ad occurrences are currently missed or overcounted, how viewership estimates vary between different 'footprints' of both ACR and STB data, and how this impacts estimates and decisions.

Jane Clarke (pictured), MD and CEO of CIMM comments: 'As more granular TV data is becoming available for attribution, we have entirely new methods and practices that need to be both vetted by the television industry and better understood by end users. This study will be a great first step for the industry to gain insight into the impact of the data on attribution results and begin to create best practices and greater industry confidence'.

Initial findings of the study will be announced at CIMM's annual membership Summit on February 6th. Web site: www.cimm-us.org .