Scale & Accuracy: Getting to Hybrid Measurement
Panels no longer suffice to get reliable estimates of fragmented impressions and viewership of niche content. There is a need to embed measurement software development kits (SDKs) directly into ads and content for passive census-based counting of exposure to streaming media. Additionally, Smart TV and STB data are providing HH or device level viewing of linear and time-shifted viewing, along with streaming data. But panels are still needed to understand who is viewing and co-viewing on TV sets. Please share your best work on:
· How are you getting to scale?
· How are you analyzing hybrid data streams (e.g., return path data from MVPDs or smart TVs blended with panel data)?
Audience-Based Buying
Digital Audience-Based media buying is now crossing into other media. Marketers are matching first party and third party data with TV viewing data from Smart TVs and STBs to conduct index-based buying for linear TV. Consortiums such as Open AP and the Advanced TV Standards Group are working to standardize audience definitions. These data are being used for planning, in-campaign optimization and ROI/attribution analysis on the back end. Please share your best work on:
· When to use age/gender targeting vs. more precise audience segments?
· Will the TV Upfront of 2019 be a watershed for reaching scale with targeted TV?
· Are large targeted ad buys getting more feasible, or do buyers still need to cobble together audiences assembled from different metrics?
· How confident are we in the quality of datasets being used for targeting, covering loyalty card data, location data, site visitation, and other behavioral data?
Attribution & ROI
Marketers increasingly want to assess campaign effectiveness using sales as the criterion of success, but don’t know how to value upper-funnel activities across the complex ecosystem. What is the right time frame for evaluating upper-funnel exposures? Is the upper funnel for branding and lower funnel for activation? Can lower-funnel actions also build brands? Is AI really being used in these models? If so, how? How do you scale time – do models looking for short-term effects devalue TV models? Many practitioners believe that good experimental design for advertising is a key missing ingredient in ROI analysis. Please share your best work on:
· What innovations can you share?
· What incrementality work have you done, what were the results and how have you helped advertisers understand their lift from advertising?
· What best practices can you share?
· How are marketers integrating MMM and both digital MTA and TV attribution?
Comparable Metrics
With traditional media, ad campaigns were planned on reach & frequency, but digital media shifted the focus to gross impressions and to outcome measures formerly associated with direct response. Reconciliation of definitions and metrics remains a critical need if cross-platform measurement and attribution are to succeed. Please share your best work on:
· What are the obstacles to overcome? How much of the concerns are related to research, rather than business considerations?
· Is duration-weighted video impressions the right metric or are there other approaches worth considering? What is the research on the impact of ad length?
· What are the potential new TV currencies to move the industry away from average commercial minute ratings? Commercial Minute Ratings? CFlight? Actual Spot Impressions?
· When considering cross-platform cross-device audience measurement, what metrics matter?
Identity & Mapping
Each consumer has multiple devices that need to be linked to unified IDs to be able to differentiate reach from frequency. Many claims are made for the efficacy of identity graphs, but can they be validated? Many have called for a next generation Universal ID in a cookieless world. Please share your best work on:
· What does the next generation of ID look like?
· How will growing demands for privacy affect these efforts?
· How reliable are methods to link individual and household devices? How can co-viewing on TV be assigned to viewing data from STBs or TV sets.
Organizational & Political Realities
Given the need to embed softward development kits (SDKs), cooperation of all parts of the distribution system is needed – but this is often lacking. Powerful players won’t carry the SDKs: some maintain proprietary systems that thwart unified views across platforms. Also, cross-media measurement is a business issue, as much as a research challenge. Sellers and buyers need to compromise on currency metrics acceptable to both sides. Please share examples of how your organization is working towards cooperation that is necessary if there were to be a Universal ID. Please share your best work on:
· How are we currently adapting to these difficulties?
· What impact is it having on media buying?
· What are points of progress?
Privacy
Increasingly, consumers take steps to opt out of measurement because of privacy concerns. What’s more, governments are taking steps (GDPR in Europe, CCPA in California) that could drastically limit the technical infrastructure supporting ad tracking and measurement. In response to these steps, some companies have redesigned browsers to reduce cooperation with cross-platform tracking while others have withdrawn permission to use their proprietary tracking systems. Please share your best work on:
· What is your organization doing to address concerns?
· What solutions do you propose?