Algorithmic multi-touch attribution (MTA) helps advertisers to accurately measure the effectiveness of their digital marketing campaigns. It is based on micro-level user data, measured mainly through technologies such as cookie-tracking or fingerprinting. It allows advertisers to understand the relative importance of individual campaign-touchpoints along the customer journey and to optimize their budget-allocation accordingly. MTA generally doesn’t include traditional or offline media, nor macro-level data such as weather.
Before the rise of digital marketing and the inherent possibilities for user journey tracking, marketing mix modeling (MMM) was the predominant method to assess marketing effectiveness. It is based on macro-level data, for example aggregated data regarding TV or print advertising spending, but also information such as unemployment rates. It’s still in use for measuring effectiveness of offline campaigns and to derive budget allocations on a strategical level. Developing robust MMM usually require many different data points per combination of dimension, which are no easily available for this kind of aggregated data. MMM fails to measure and assess the performance of specific marketing programs and campaigns on a tactical or operational level, simply because the data being used is not detailed enough.
Working with MTA and MMM side by side isn’t trivial, since they work on different levels of data and might even produce contrasting insights. Instead of forcing these two approaches to work together, the aim is to move toward a unified marketing model. Forrester Research calls such a unified marketing impact analytics (UMIA) approach as “A blend of statistical techniques that assigns business value to each element of the marketing mix at both a strategic and tactical level.”
This get marketing managers closer to the single point of marketing truth. According to Forrester “the integration of online and offline data, to measure short- and long-term goals, gives marketers both allocation and planning capabilities in one tool”. UMIA facilitates evidence-based budget allocation, measurement of cross-channel effects and detailed user-journey insights.
A key requirement is integrating customer-level data, such as cookie-level user journey tracking and aggregate-level media performance data, such as TV GRPs. Cross-device tracking is key, i.e. connecting a user’s marketing touches across devices, for example a campaign click on mobile device with a purchase in the laptop-browser.
Regarding the methodical perspective, at Adtriba micro-level user data and macro-level aggregate data (e.g. sunshine duration at a specific geo-location) are integrated in the same underlying logit model. This allows to easily quantify the impact of aggregated data on a particular customer journey as well as on all sales over time.
Of course UMIA also is a new hype that might allow consultants and respective solution vendors to create new business opportunities. Many marketing analysts and data scientists probably have already worked with this approach without calling it UMIA.
Also, not for every company the effort of gathering and integrating macro-level data makes sense. A young e-Commerce startup might not even have the historic data available to conduct UMIA. Instead it should focus on getting the algorithmic multi-touch-attribution part right with a simple and cost-effective solution, to improve operational and tactical marketing effectiveness.
Nevertheless, the essence of UMIA is tear down the walls between classical MMM and digital driven MTA, to bring aggregated data and user-level data together. Integrating these two level of data will drive better marketing decisions and improve performance in a cross-channel and cross-device perspective.