Home AdExchanger Talks There’s No Such Thing As An Attribution Easy Button

There’s No Such Thing As An Attribution Easy Button

SHARE:
Madan Bharadwaj, founder, M^2

There are a lot of misconceptions about ad measurement.

Some marketers believe media mix modeling is superior to multi-touch attribution or that it’s possible to pinpoint the exact impact of each marketing touch point on sales. (Neither is true.)

But the biggest thing most marketers are wrongheaded about is in thinking there’s a single solution for attribution.

It simply doesn’t exist, says attribution expert Madan Bharadwaj, on this week’s episode of AdExchanger Talks.

“There is no one technique to rule them all,” says Bharadwaj, who founded M^2 (pronounced M-Squared) in 2022, a measurement startup focused on advanced marketing attribution solutions.

Before that, Bharadwaj co-founded media effectiveness startup Measured and was VP of product management at attribution vendor Visual IQ before the Nielsen acquisition in 2017, where he helped launch the first algorithmic MTA product.

“You have to bring [multiple approaches] together,” he says. “The magic or the mystery is in how it comes together for your brand.”

There’s no single method that can account for all factors because, fundamentally, measuring the effectiveness of advertising is about measuring the complexity of human behavior.

“We can only understand human behavior to a certain level of precision,” Bharadwaj says.

If, for example, you measure the effect of media on behavioral patterns through an incrementality test or if you make estimations through a regression model, like MMM, you will get similar answers, Bharadwaj says, “but not exactly the same answer.”

Which is why marketers should triangulate measurements from multiple sources before making any single investment decision.

“That is an important concept that’s slowly beginning to take root in the industry,” Bharadwaj says, “that there is no one source of truth.”

Also in this episode: Base attribution vs. advanced attribution, whether marketers can trust platform-provided measurement (sort of, not really), how to accurately benchmark media channels. Also: Spitballing attribution-themed cocktail names. (Anyone care for a refreshing “Incrementality on the Beach”?)

For more articles featuring Madan Bharadwaj, click here.

Must Read

The Largest Shopping Mall Operator Has Its Own Retail Media Network

Simon Property Group, the largest shopping mall operator in the world, is taking its biggest step yet into the world of data-driven advertising. On Thursday, the company launched Simon Media Network, its version of a retail media network.

The Trade Desk’s Zuma Update Adds A Host Of AI-Powered Easy Buttons To Its Kokai Platform

TTD is introducing agentic AI workflow improvements and automated audience building, which are quickly becoming table stakes for programmatic platforms.

NBCU’s Streaming Strategy Involves Revisiting The Cable Playbook

In the modern streaming landscape, everything old is new again. At least that’s the case for NBCUniversal, which is trying to bring back bundled distribution deals and appointment viewing.

Privacy! Commerce! Connected TV! Read all about it. Subscribe to AdExchanger Newsletters
Joel Meyer, Chairman, Prebid.org

New Chairman Joel Meyer Dishes On Prebid’s Reset For The Agentic Ad Tech Era

OpenX CTO Joel Meyer, who was named Prebid’s new chairman, previews the org’s next phase and helping publishers navigate competing agentic protocols.

These are the days of our lives

After Years Of Fighting LGE, Alphonso Is Headed For An IPO – Unless Comcast Or The Koch Brothers Get There First

Alphonso – LG’s ad tech division – finally has a way out of its legal drama with parent company LG Electronics. Actually, it has three potential options.

Micro1 Wants Human Domain Experts To Profit From AI And LLMs

Much like the ecosystem of life that surrounds a blue whale, a market of AI SaaS vendors is springing up around the biggest AI companies. And AI data startup Micro1 is emblematic of the shifting nature of these early-stage AI vendors.