Big data keeps getting bigger as the number of devices we use on a daily basis continues to grow—and it’s confounding marketers looking for the most accurate measurement of their marketing return on investment (ROI). But instead of a single measure, a multi-model approach may be marketers’ best chance to measure the full impact of their marketing effectiveness.
Nielsen’s president of global MROI solutions, Ross Link, recently shared findings from the Digital Media Consortium (DMC) II at Re!Think 2016 about how such an approach can help fill the gaps inherent in big data today. However, this strategy isn’t one size fits all—the type of data a marketer has available in terms of granularity and coverage will dictate what methodology they should use.
Nielsen founded the DMC in response to the recent proliferation of big data that has driven a rapid change in marketing ROI measurement techniques. The goal of the DMC II was to improve industry practice in the precise measurement of digital media effectiveness. Specifically, it used granular, household-level data to see how common statistical methodologies with different data limitations—marketing mix modeling (MMM), multi-touch attribution (MTA) and matched panel analysis (MPA)—can work together to get the most accurate ROI for marketers.
“We thought one modeling methodology would come out on top, but it turns out that each approach still deserves a place in the modern marketer’s toolkit. When to use MMM, MTA or MPA depends on the type of data a marketer has available and what they’re ultimately looking to measure,” said Link. “For example, when the impact of all marketing can be tied to a particular individual (i.e., single-source data), the ideal solution is hands down household-level regression (MTA). This may be possible right now for marketers who invest in only digital tactics, but for the majority of brands that leverage offline marketing vehicles today, single-source data likely won’t be available for at least another several years in the U.S. and for decades to come in developing markets.”
Link continued, revealing that until single-source data is a reality, marketers should leverage store-level MMM, particularly when looking to optimize ROI holistically across all business drivers—both online and offline. On the other hand, when the goal is to focus solely on measuring ROI for digital-specific media investments like display and video, marketers should leverage household-level MPA.
DMC I included a partnership between Nielsen, Google, Facebook and seven major advertisers to identify groundbreaking insights on accurately measuring the ROI of social (including paid, owned and earned digital media). The study was based on 1.5 billion Facebook impressions, 600 million Google search impressions and an analysis of $2 billion in sales.
DMC II was formed to solve attribution analytics challenges in the industry given data available today. DMC II analyzed 11 consumer packaged goods (CPG) brands across nine categories that encompass $30 billion in sales, 3.6 billion digital display impressions, 300 million digital video impressions and 4,000 advertising campaigns.