NEW YORK: Marketers who rely on cookies to understand consumers are effectively using "broken" sources of insight, according to an executive from Atlas, a measurement service owned by Facebook.
James Dailey, Atlas' product marketing manager, discussed this subject at the Advertising Research Foundation's (ARF) 2015 Audience Measurement conference.
"Some of the very fundamental metrics with which you manage your digital advertising are broken," he warned. (For more, including additional data, read Warc's exclusive report: How people - not cookies - can define digital audiences.)
By way of an example, he cited Atlas research showing that figures regarding the reach of digital ads are generally inaccurate.
Extrapolating this trend outwards for campaigns which are seeking to secure the attention of large numbers of people brings greater clarity to such a problem.
"The data for some of our larger clients," Dailey reported, "showed they're reaching more people in California than live in California [and] more people in New York than live in New York. "At that superficial level, the data are clearly broken."
Frequency, by contrast, is typically "understated", meaning brands are not getting a realistic picture about how often individuals see their messages.
"You think you've capped a demand-side platform buy at ten impressions [but] you've actually served more like 24," said Dailey.
The main cause of this failure is that cookies reside on separate devices and browsers, but cannot connect these various gadgets and services.
Consequently, while individuals usually visit their favourite websites via multiple routes, cookies are unable to determine that fact.
"A person exists across browsers; a person clears cookies; a person uses mobile devices," Dailey said.
Solving this issue will thus depend on creating systems able to identify specific consumers wherever they are. "The industry is at the point of a major transition," Dailey reported, "from cookies to people."
Data sourced from Warc