In recent years various leaders and marketers have pronounced multi-touch attribution (MTA) as dead. Each cites different reasons--the loss of signal from identity deprecation, failed experiments, or a fallback to MMM frameworks. However, despite these proclamations, MTA is steadily adopted by a significant percentage of businesses each year. So what's the problem?
Below is a summary of “MTA is Dead, Long Live MTA,” an MMA-sponsored research paper about the past, current, and future position of MTA within measurement frameworks. The Mobile Marketing Association (MMA) recognized MTA as marketing’s best form of measurement in 2016. Given the slow adoption success and general threats to user-event data, MMA researched and reassessed the state of MTA in December 2021. Their verdict: MTA fundamentals are as strong as when first begun over 15 years ago. Moreover, the progress of addressable digital TV, practical privacy preservation techniques, and identity resolution might make the solution even stronger than originally implemented. Considering both the challenges and benefits, MTA should maintain a key position in the marketing analytics toolkit.
The value of multi-touch attribution (MTA) or full-path attribution is undeniable to its practitioners. Brands can calculate channel efficacy and efficiency, understand the customer journey in aggregate and discrete individual paths, and identify the most valuable sources, leads, and marketing elements. The resulting optimization opportunities cross all marketing elements: tailoring for segments, optimizing content creation and distribution, improving campaign performance, and aligning to business goals. Attribution drives optimization that increases revenue.
Various leaders and practitioners have announced the death of MTA in recent years. And while the rate of marketers using or planning to use MTA has remained nearly constant, it has not grown. For example, since 2016 nearly 40% were using MTA, with another 30% planning near-term implementation. But in 2021, the numbers are nearly the same. MTA adoptions fail to grow, yet the same percentage is trying to implement each year.
Challenges to user-level tracking are numerous and growing, but the recent restrictions on IDFA, regulatory realities (GDPR, CCA, etc) and the imminent loss of 3rd party cookies are often mentioned in the same breath as MTA’s death. While these create challenges for traditional user-event tracking, they are not unassailable obstacles for newer methodologies, nor are they the reason many brands have false starts or even failures implementing MTA.
The greatest obstacles include data, but not data alone. Most respondents reported data-compounding obstacles such as mismatched legacy measurement methodologies or organizational inertia. However, in each case, these same obstacles would block any broad contemporary or future-oriented measurement framework. And each challenge is more internally-driven than the result of external environmental factors such as IDFA, GDPR, or 3rd party cookies.
Walled gardens are usually the first and obvious challenge for brands with a complex marketing mix, yet other barriers center on the fundamentals: data quality, access, and linkage. Many brands don’t trust their own data. CRM data includes 20-40% duplicate entries, and in most cases, not all employees can access owned data (much less walled-garden data). Few report having an ability to connect between different platforms or to centralize customer data in any meaningful way. A majority of senior marketers felt their data was insufficient in its current form to drive decisions. And each new channel added to the marketing mix threatens to add a new data silo.
Success in a data-driven (and ultimately, AI-driven) world demands brands increase their data science maturity, and that means getting a handle on all their data sources, platforms, and silos. While these data challenges will prohibit the development of full-path attribution, they will also doom any comprehensive measurement framework that seeks to measure across the brand, understand the complete customer journey, find and optimize (or personalize) the most important touchpoints, and make allocation decisions.
Legacy measurement, even when done well, can cause unintentional barriers to MTA success. Historical measurements such as MMM or specific point solutions use different approaches, different inputs and outputs, and don’t align with MTA in timelines or outcome metrics. Often, the ownership of legacy approaches naturally includes self-interest. Rather than a fading challenge, mismatch of varying measurement methodologies will likely grow. Of the marketers surveyed or interviewed, about half are interested in integrating MMM and MTA.
The solution is not which methodology to use, but how to get the benefit of both: measure brand strength and campaign efficacy, brand awareness and specific channel efficiencies. An effective framework will need multiple approaches, each with unique insights from different methodologies. Collaborative inertia is strong, and falls squarely on the practitioners to make inroads and a case that values the best application and outputs of different measures. Ultimately, a “winning framework” may be an intentional mix of MTA, MMM, and experimentation across and beyond both methodologies.
More than two-thirds of respondents found organizational and leadership support a key challenge with implementation success. In addition to channel- and methodology-specific silos to address, marketers are now challenged by legal departments over data privacy and security, by IT departments concerned by martec stack complexity, and even operations teams that own specific customer data. Corporate functions are at odds with the analytics teams. Unfortunately, support for change–process, tech, cross-functional–has decreased in recent years.
Gaining support from leadership includes proving the value of different inputs and outputs, each measuring different parts of the marketing and customer value. Until trust is generated to break down walls and demonstrate a complete measurement picture of the customer journey and user-events, marketers will fail to create holistic frameworks that capture the full opportunity across the brand’s value proposition.
Recent developments in addressable TV (OTT, AVOD, FAST, etc.) are opening up new accessible channels for MTA. Privacy preserving practices, including clean rooms and privatized measurement is increasing first-party data value. Identity is no longer locked on specific “hard” offline identifiers (emails, phone numbers) or digital ones (cookies, IDFA). Instead, identity as a framework is not a singular technology solution, but an approach that can achieve identity resolution.
MTA or not, brands must get a handle on their data, create complimentary methodology paths, and demonstrate value across corporate functions and leadership. Regardless of MTA adoption, centralization of clean, accessible customer data drives organizational power across both customer-facing and strategic functions. Data volume and channel variety growth are accelerating, not stabilizing. And whether MTA is part of the analytics toolkit (it should be), allowing the continued siloing of methodologies will never achieve synergies or the insights possible from integration.
MTA remains one of the best tools to measure outcomes. For example, Arcalea's Galileo captures the complete full-funnel buyer journey, identifies high-value tactics and assets, and increases revenue and efficiencies through reallocation and optimization. To align business to take advantage of its inherent value means preparing your business for the future of big data (and AI), varied methodologies, and numerous touchpoint opportunities. Waiting for a perfect solution, whether a fully integrated MMM-MTA or a universal identifier, is a recipe for brand stagnation.
If you let cookie deprecation kill your ambitions, you didn’t dream big enough.—Eddie Drake, SVP Marketing, Bank of America
Focus on data quality. Assess the data environment, ensure that all parts of the data organization, infrastructure, and governance meet immediate and future needs. Inventory analytics processes across the organization. Methodologies will need to be diverse (and verifiable) to continue capturing new channels without creating new silos. Practice agile principles of responsiveness: implement, learn, experiment, extract and apply insights in real time. Organizational inertia will kill far more than just a measurement framework. Brands that promote data hoarding and private, unverifiable point solutions are inviting dysfunctional decision-making in which self-preservation is a stronger driver than outcomes.
Within the fluid environment of marketing, MTA as envisioned in the early 2000s may no longer be the model. However, MTA’s power of describing the customer journey, the most impactful touchpoints where customers make decisions, and ability to tell the story is finding new methods to provide even more powerful advantages.
Customers want a personalized experience, and brands want to measure specific impacts.
Waiting for all the MTA challenges to generate a perfect model is a fool’s errand. Solutions will continuously evolve. And the brands who activate the benefits of MTA earliest have the advantage. MTA is refusing to die regardless of specific brand setbacks because of the advantages it creates.