Meghan Corroon of Clerdata: Why Statistical Modeling — Not Attribution — Is a Better Way to Measure Marketing ROI

“We're really disrupting this and saying, ‘data science is blowing through the roof right now across a lot of levels, so we can do better.’” — Meghan Corroon, CEO, Clerdata

Is attribution dead? According to Meghan Corroon, it’s not quite the best way to measure marketing effectiveness — if it ever was to begin with.

Corroon is the CEO of Clerdata, a marketing intelligence platform that uses statistical modeling to understand the effects of marketing, retail media, and trade actions on a brand’s bottom line. 

Corroon joined a recent episode of the “Unpacking the Digital Shelf” podcast, “Attribution is Dead. Long Live Statistical Modeling,” to share why a science-based approach can help brands better understand what they’re actually getting in return for their marketing dollars. 

The Problem With Attribution 

For nearly three decades, brands relied on cookies to understand consumer behavior and effectively target customers. But the death of the cookie has upended digital advertising, making attribution less reliable. 

“The ability to track a person across the internet has been vastly deprecated,” Corroon says. 

Brands have had to figure out how to navigate a cookie-less world. Corroon says many of them still use attribution as part of their measurement strategy, even though they don’t fully believe in it anymore. 

Clerdata offers them an alternative that’s more accurate than attribution: statistical modeling. 

The Clerdata Way

Corroon says precision and timeliness are the two most important characteristics brands need to truly understand their return on investment (ROI) and make more informed marketing decisions.

Clerdata has built these characteristics directly into its platform. The solution uses statistical models to deliver an accurate, real-time picture of incremental ROI. The picture encompasses sales dollars the brand wouldn’t have won without a specific sales tactic and analyzes whether the tactic was ultimately cost-effective for the brand.

The model uses a company’s historical data and current sales data to understand the impact of marketing, retail media, and trade spend on incremental sales. Clerdata also helps brands understand halo effects or the positive network effects of a tactic on incremental sales. 

With this approach, brands can become more agile and quickly make changes that drive top-line revenue growth and profitability. 

During a time when marketing budgets are flat or even declining, brands need to drive more value from their spend. Clerdata integrates an evidence-based approach into a powerful software platform, delivering insights it previously would’ve taken years for brands to get from expensive consultants performing a mixed-marketing analysis. 

Clerdata addresses this challenge while giving brands an effective alternative for attribution.

“We're really disrupting this and saying, ‘Data science is blowing through the roof right now across a lot of levels, so we can do better,’” Corroon says. “Business models of those companies are what they are, but we're agile, we're disruptive, and we're saying, ‘No, it can be better, more rigorous, and it can be much more real-time so that your business can make better decisions.”

Understanding Incremental ROI and Halo Effects

With Clerdata, brands can more effectively assess the performance of various tactics, including earned and owned media, organic social media, retail media networks, and marketplace channels.

Corroon says Clerdata can also do an apples-to-apples comparison of how one channel drives sales versus another, such as comparing Kroger 84.51° to Google ads. Additionally, the platform can conduct complex analyses that provide a holistic view of incremental sales and how investments in one channel ultimately lead to customer conversions in another.

Corroon gave the example of a consumer searching on Amazon for a product, putting it in their basket, but never checking out. The brand then retargets that consumer across other digital channels, reaching them with a mobile ad when they’re in Target, which prompts them to buy the product in-store rather than on Amazon. Corroon refers to this as a halo effect. 

“It's not just what they're selling on Amazon, it's also affecting consumers' behavioral choices in other settings where they might buy something.” — Meghan Corroon, CEO, Clerdata

Delivering Results for Brands With Statistical Modeling

Many of Clerdata’s customers have discovered that the tactics they thought delivered high ROI actually don’t. Others have found that they’ve underestimated the impact of certain channels, such as earned and owned media or organic social media. 

“We're going to see more and more granular tweaking and optimizing of marketing strategies, but right now they're shifting major pockets of money around because there's a lot of stuff that has been assumed to have been working that's frankly not working very well,” Corroon says.

Many of Clerdata’s customers have seen anywhere from a 20 to 35% marketing efficiency gain. While Corroon touts the effectiveness of her platform, the fact that these tactics have historically been poorly measured also contributes to these gains, she says. 

With statistical modeling, brands may finally be able to put attribution in the rearview and generate a greater return on their marketing investments. 

“The reliability of that old school way of doing it, the attribution approach, any ROI on that kind of tracking mechanism is basically gone,” Corroon says.

Listen to the full episode to hear more of Corroon’s insights on statistical modeling.