A Multi-Touch Attribution Model in a Complicated World

Finding the ultimate model for Multi-Touch Attribution (MTA) is difficult if not impossible but the Mobile Marketing Association (MMA) believes that, while not perfect, their newly released MTA model is a big step forward in leveraging data to ascertain attribution.  At a recent breakfast during Advertising Week, Greg Stuart, Chief Executive Officer, Mobile Marketing Association, hailed it as “the next generation measurement initiative which gives user-level data proportional credit against a granular list of touch points.”

The path to a better MTA model was not easy.  When it comes to attribution there are many unknowns in the consumer journey.  At what point was the sales decision made?  Was it a banner ad that flipped the switch in consumers’ minds?  Was it a TV spot or maybe a recommendation from a trusted friend?  In addition to these unknowns, the consumer consideration process varies by category and even, according to Stuart, by platform within companies.  Buying toothpaste has a different purchase cycle than buying a car.  Buying in a Walmart brick-and-mortar store is different from buying through Walmart.com.

”Attribution accuracy largely depends on the quantity and quality of the data,” noted Antonio Lyon, CEO U-Glove, who attended the breakfast.  Key factors in the consumer’s buying journey are excluded from the data, such as word of mouth, or brand affinity, so that it doesn’t tell the whole story.”

The MMA’s MTA model does take a deep dive into a range of datasets and is highlighted in a data map that takes into consideration four buckets regarding data: Linkability, Potential Aggregation, Uses in Profiling and Making Conversions.  Starting at the core, the MTA relies on unified user data IDs that bring together different devices and customer interactions -- then adding advertising cost data in order to calculate ROI, main data categories such as linkable marketing, aggregated datasets, conversions and audience segmentations, technology for linking all of the types of data whether cookies, IDs or GPS, data delineated by sub-types such as website visits and sales and finally, sub-types of data by platform such as desktop and mobile.

MMA’s model might be another step in the long road to multi-touch attribution.  CIMM recently commissioned a study of attribution conducted by Sequent Partners and concluded that, as an industry, we have made some progress towards the development of an attribution model across media. But we still have a lot of work to do and, at least in the near term, there are few viable systems currently available that holistically capture cross-platform consumer behaviors.

“The industry is too focused on upstream processes, mainly driven by the overwhelming growth of marketing technology, such as AI, at the expense of more traditional, proven ways to truly connect with the customer,” Lyon concluded.

This article first appeared in www.MediaVillage.com