Marketing-mix models face a variety of challenges in achieving true accuracy, including the prevalence of walled gardens and the challenges posed by social media, a paper in the Journal of Advertising Research (JAR) has argued.
The study, entitled How limited data access constrains marketing-mix analytical efforts: why data barriers are preventing marketers from optimizing marketing spend, was written by Gian M. Fulgoni, the former chairman/CEO of comScore.
“Marketing-mix models fundamentally are designed to measure the impact on brand sales of each of the key elements of a brand’s marketing mix,” he wrote.
“Broadly, marketing-mix modeling consists of statistical analyses, such as multivariate regressions, using sales and marketing time-series data to estimate the impact of various marketing tactics on sales.”
Or so the theory proposes. In practice, he said: “Walled gardens, such as Amazon, Facebook and Google retain their generated data as first-party property, limiting client firms.”
The European Union’s General Data Privacy Regulation (GDPR) and an upcoming data-privacy bill in California pose further constraints, too.
Again, in theory, Fulgoni writes: “Due to the increase in cross-platform media, marketers need single-source models to isolate accurately the effects of the many touchpoints.”
But social media further undercuts any kind of marketing-mix modeling: “Facebook alone reported having 1.47 billion daily active global users in the second quarter of 2018,” Fulgoni asserted.
“When the use of social media first reached critical mass, researchers hoped that social media data would be able to predict offline brand communications.
“A result, they assumed, paid, owned, and earned social media usage would create digital data that would be of substantial value in marketing-mix modeling. Research has shown, however, that this is simply not the case.”
Fulgoni concludes: “The online and the offline world behave like different ecosystems, reducing the use of social media to marketers.
“Going forward, marketers can either run marketing-mix models that exclude the causal components for which data are unavailable, or they can use causal data that are aggregated across users.
“It is very likely that the media industry will have to figure out how to appeal to consumers for their data at scale – and what to provide in exchange. The era of passive measurement, without consumers’ knowledge and permission, may be coming to a close.”
The study, entitled How limited data access constrains marketing-mix analytical efforts: why data barriers are preventing marketers from optimizing marketing spend, was written by Gian M. Fulgoni, the former chairman/CEO of comScore.
“Marketing-mix models fundamentally are designed to measure the impact on brand sales of each of the key elements of a brand’s marketing mix,” he wrote.
“Broadly, marketing-mix modeling consists of statistical analyses, such as multivariate regressions, using sales and marketing time-series data to estimate the impact of various marketing tactics on sales.”
Or so the theory proposes. In practice, he said: “Walled gardens, such as Amazon, Facebook and Google retain their generated data as first-party property, limiting client firms.”
The European Union’s General Data Privacy Regulation (GDPR) and an upcoming data-privacy bill in California pose further constraints, too.
Again, in theory, Fulgoni writes: “Due to the increase in cross-platform media, marketers need single-source models to isolate accurately the effects of the many touchpoints.”
But social media further undercuts any kind of marketing-mix modeling: “Facebook alone reported having 1.47 billion daily active global users in the second quarter of 2018,” Fulgoni asserted.
“When the use of social media first reached critical mass, researchers hoped that social media data would be able to predict offline brand communications.
“A result, they assumed, paid, owned, and earned social media usage would create digital data that would be of substantial value in marketing-mix modeling. Research has shown, however, that this is simply not the case.”
Fulgoni concludes: “The online and the offline world behave like different ecosystems, reducing the use of social media to marketers.
“Going forward, marketers can either run marketing-mix models that exclude the causal components for which data are unavailable, or they can use causal data that are aggregated across users.
“It is very likely that the media industry will have to figure out how to appeal to consumers for their data at scale – and what to provide in exchange. The era of passive measurement, without consumers’ knowledge and permission, may be coming to a close.”
Sourced from JAR
