More Marketers Are Adopting Multi-Touch Attribution, But There’s Still Some Frustration

Multi-touch attribution is reaching an inflection point (finally).

Multi-touch attribution is reaching an inflection point (finally).

The majority of large advertisers (53%) now say they use MTA to track and optimize their spending across channels, according to new research from marketing trade org MMA Global, released on Monday.

That’s the highest peak since the MMA, which represents more than 800 marketers, mar tech companies and publishers, started tracking MTA adoption in 2016.

On the fast track

But why now? MTA has been around for years.

The pandemic could have had something to do with it, said Brad Feinberg, North America VP of media and consumer engagement at Molson Coors, which began experimenting with MTA around five years ago and now has a fairly robust MTA-based system in place through a partnership with IRI.

Looking at past performance to inform future decisions isn’t helpful when the world has changed and historical reference points no longer apply, said Feinberg, who also chairs the MMA’s media and data board.

But even without an event as disruptive as the pandemic, the world of media in general changes so rapidly now that historical models based on long-term behaviors “aren’t accurate enough” to make predictions, he said.

“I mean, as of three months ago, Netflix didn’t even exist in the advertising space,” Feinberg said. “We need tools that can account for change when we start to invest in these places.”

Even so, it’s not as if marketers just woke up this year and decided to embrace MTA. There’s been “a steady movement over time,” said George Wu, director of advanced analytics at AT&T, which also has an MTA program in place.

“This is part of a broader movement towards using analytics to make marketing decisions,” Wu said. “I’d imagine that in the not-too-distant future, pretty much any company that runs digital marketing will have some form of MTA.”

Comic: The Attribution GameMTA vs. MMM

In the simplest terms – although setting up a multi-touch attribution program is far from simple – an MTA model measures the effectiveness of advertising by assigning fractional credit to touch points in a customer journey.

Using MTA, marketers can get a more nuanced, quicker read on what’s working and what isn’t so they can be more flexible with their optimization.

But MTA requires access to individual-level data tied to trackable conversion events, and that’s getting scarcer.

In light of signal loss, some marketers have been turning to marketing mix modeling (MMM) to analyze their marketing spend. Amex, for example, started a project in 2019 to replace MTA with an MMM-based approach in anticipation of the end of third-party cookies in Chrome.

MMM uses statistical analysis to measure all of the different factors that can influence a sale – seasonality, market trends, promotions and other marketing-related activities – to try and determine the relationship between them.

It’s an aggregated model, which means it’s not reliant on identifiers to do its thing. But MMM does require access to a heck of a lot of historical data, often at least two or three years’ worth.

“Super-obvious statement here, but if you’re a marketer, it’s probably important to you to be able to quickly follow consumer trends,” said Greg Stuart, CEO of MMA Global. “The challenge with MMM is that it doesn’t pick up signal fast enough as new things are happening, but MTA does because it’s real-time measurement.”

Signal loss

Problem is, addressability is getting a lot harder. Apple’s AppTrackingTransparency framework is only the start.

It’ll be up to the industry to innovate, said AT&T’s Wu. Companies like Meta, Nielsen and Neustar are laser-focused on developing new solutions … although it’s not clear if they’ll be viable before Google’s third-party cookie deadline in Chrome hits at the end of next year.

Comic: Mount Cookie“Advertisers and companies that use MTA are watching the landscape very closely, because next year is going to be critical for MTA,” Wu said. “But I can see a world where a lot of advertisers drop MTA next year and then pick it up again in a couple of years when the solutions have matured.”

But signal loss doesn’t have to derail an MTA program, Stuart said.

“There’s some negativity about MTA because of what’s going to happen with cookies and because of what I’d call ‘walled garden behavior,’ but while we might be losing signal in some channels, we’re gaining signal in others,” he said. “Take TV – it’s going digital at a rate that makes your head spin.”

Worth the squeeze?

But even if audience addressability weren’t getting more difficult, MTA always has been (more difficult, that is).

Marketers need to organize their data properly before they even get started and must be ready to make ongoing investments in time, effort and money. There is no magic switch to flip, and for a long time that “wasn’t fully clear to people,” Stuart said.

Despite the fact that 53% of marketers are using MTA, according to the MMA – and 27% have reached full deployment – the overall Net Promoter Score for MTA solution providers is … still in the toilet, although it’s getting better as marketers align their expectations with reality.

In 2016, the overall NPS for MTA solutions was minus 42. Now, it’s down to minus 16.

“People went into this with a sense of excitement about analytics and transforming their marketing, but they hadn’t done the groundwork,” Stuart said. “You have to do the groundwork.”

Like with anything, you get out what you put in.

Molson Coors, for example, is seeing its multi-year investment in MTA pay off, Feinberg said. It’s able to measure the impact of its marketing on sales all the way down to a specific piece of creative or certain copy within a campaign, which helps drive down the cost of production.

There’s a level of granularity you can get with MTA that MMM can’t provide, Feinberg said – although there can be value in a hybrid approach.

“I see marketing mix modeling as like a double check on your MTA results,” Feinberg said. “MMM, when done well, will help you with the really big macro ROI-type metrics, but it hasn’t given us enough granularity to understand what’s really moving and working well.”

Stuart pointed to a recent campaign from Wendy’s to illustrate the point. The purpose of the campaign was to drive in-store traffic. Using an MTA-based approach, Wendy’s found that two segments in particular – people who use coupons and people who drive by a Wendy’s location – performed five times better than any other tactic in the campaign. Each of those segments represented 5% of the budget.

“If you can take 10% of your budget and make it work five times better than anything else you’re doing? You don’t have to be a math genius to see the value,” Stuart said. “MTA tells me where to focus – and then you can just iterate the hell out of it.”

Enjoying this content?

Sign up to be an AdExchanger Member today and get unlimited access to articles like this, plus proprietary data and research, conference discounts, on-demand access to event content, and more!

Join Today!