How GroupM’s inclusion list decides which publishers it sells programmatically

Enter GroupM’s m-List, an “inclusion list” only available in the U.K. of about 6,000 domains that have been hand-selected through a combination of criteria based on verification firms like DoubleVerify, Integral Ad Science and Jounce that assess viewability and fraud risk, as well as manual validation from a committee of m-List overseers who check to see if the sites on the m-list are visited by actual humans and aren’t MFAs.

What is the m-List?

GroupM’s team in the U.K. introduced the first iteration of the m-List three years ago, implementing it across all of the U.K. division’s programmatic inventory. Dave Sandham, GroupM’s engagement and operations director, said the company first set out to clean up the programmatic supply chain about six years ago.

Over time, the m-List grew to run across 95% of all GroupM’s programmatic campaigns in the U.K., including all of the programmatic campaigns sold through the company’s programmatic firms Xaxis and Finecast.

“There’s a few exceptions every now and then when a client says, ‘I really want to run on these domains,’ and it’s normally because we haven’t vetted them for the m-List or they’ve got a relationship with that partner,” said Sandham.

The list wasn’t a set-it-and-forget-it initiative. At least once per quarter, a committee of GroupM programmatic specialists review the roughly 6,000 domains (coming from just 50-60 publishers that GroupM has trading relationships with, though Sandham declined to name them) to see if they still meet the criteria to be on the list and vet any new potential domains.

What is m-List’s criteria?

GroupM’s criteria is based on fraud and brand safety, but that criteria gets updated every three to six months to stay current with industry benchmarks, Sandham said. Criteria ranges are set on ad viewability and ad clutter, with the current viewability threshold is between 60-80%. For example, this range has been raised over time, Sandham said, because viewability has improved holistically across the internet.

If any of the m-list domains fall below these criteria, however, they aren’t automatically removed from the list. Instead, Sandham said his team will consult with that publication to help them improve their metrics.

The human verification, performed by a committee member, assesses whether sites are measuring real traffic by analyzing variables that include legitimate contact information for a publisher and what the traffic patterns to the website are like through partners like SimilarWeb.

“We have to be conscious that you don’t have those sites gaming the system to increase the viewability at the expense of content and the user, which is why we have that additional manual check on top. [We also do] MFA checks against external providers as well,” Sandham said.

Where sustainability fits into the equation

The m-List inclusion does not measure sustainability due to the ad industry’s measurement of carbon emission still being in a rudimentary state, Sandham said, though early assessing is being done in partnership with carbon emission measurement firm Scope3. Eventually, sustainability will be a benchmark to be on the list, but Sandham did not provide an exact timeline.

Sandham countered that the list does already lessen carbon emissions by reducing the domains allowed in its programmatic business. 

How has the m-List changed the performance of programmatic advertising? 

For publishers, the average CPM for inventory sold through the m-list was double the amount of any other programmatic campaign segment, according to a LinkedIn post by Adam Walsh, head of media innovation at GroupM. Meanwhile, GroupM’s clients saw conversion rates that were four times higher than comparable campaigns not run on the m-list, leading to a 43% lower cost-per-acquisition, he added.

“I’ve lost count of the number of times I’ve heard variations of, ‘It’s a performance campaign. My client just wants cheap inventory.’ … [The results show] a clear indication that the ‘quality’ conversation shouldn’t be limited to Brand [sic] campaigns,” Walsh argued. 

Is this the solution publishers are looking for? 

Strategies like the m-List on the sales-side aren’t going to be the only solution improving domain curation within programmatic. And not all publishers might be convinced that this solution makes sense — especially as the list is only available in the U.K. without concrete plans to offer a similar list globally.

When inclusion lists become too narrow, many publishers run the risk of being excluded entirely, and without one set definition of what an MFA site is, the risk of being dubbed an unqualified domain increases. 

“Not every article on every publisher is going to be very serious, it’s not a perfect ecosystem. But if [an advertiser] only buys the [inventory] that [is deemed] good, the question becomes, is the ecosystem big enough for the kind of scale and reach that people want?” said Jana Meron, who serves as a strategic advisor for programmatic and data strategy at her consultancy company Lioness Strategies.