How Attribution Can Drive Marketing Strategy
One of the most enduring questions about the resource allocation concerns the proper mix of online and offline advertising investment. What is the right mix of ad spend between TV and digital? Answering that question often involves applying a classic bit of marketing effectiveness analysis known as marketing attribution. Essentially linking a business outcome to the marketing initiative that drove it, attribution analysis is not only integral to advertising success but also a key driver of securing buy-in from the executive boardroom.
Until very recently, calibrating online and offline advertising investments involved marketing mix modeling (MMM) or econometrics. This type of high-level, probabilistic attribution has long been the go-to model for yielding valuable insights regarding the relative effectiveness different advertising channels have on consumer behavior. With that knowledge, marketers are able to forecast future business performance and adjust their budgets to achieve better outcomes. Two drawbacks of MMM, however, are that it’s typical ‘backward-looking’ and that it takes significant time to execute.
The advent of digital media has brought an onslaught of a real-time user and event-level data that allows marketers to attribute an individual’s behavior to highly customized marketing levers. Because these insights are immediately after (and sometimes during) the campaign, marketing channels and tactics can be optimized to generate maximum value while the campaign is actually running, ensuring the maximum possible return on investment.
Although real-time, in campaign data and optimization offer powerful incentives, marketers must include Marketing Mix Modelling in their analytical toolkit. MMM provides a 30,000-foot-view that cannot be obtained from the event-level insights offered by user-level analysis. MMM also provides a long-term view of marketing effectiveness the can illustrate the benefits of brand building, which translates into long-term brand returns.
The truth is that the most effective analytic strategies combine a channel-level, pre/post campaign view with event-level insights revealed by digital data. Below I have detailed four important steps to creating an analytical toolkit:
۱. Build A Data Infrastructure
Precious few companies have the data infrastructure needed to sift through the never-ending stream of digital data, nor the expertise needed to mine for the golden nuggets. Companies that make an upfront investment in a strong, holistic data analysis capability will stand to gain immeasurably from the more effective marketing campaigns they will generate. Remember that user-level data works in concert with more traditional analysis, not in lieu of it.
۲. Ask, “What is the problem we’re trying to solve?”
This is the question every marketer should ask before spending a single dollar. Without an answer, all the data in the world won’t save brands from pursuing costly endeavors of questionable value. How can success be measured when it hasn’t been defined? Whether it’s increasing brand health, generating sales, building long-term value or any other goal, companies need to reach consensus on the key marketing performance objectives they are attempting to achieve and then activate their attribution capability to help them build out the right strategy and tactics.
۳. Audit and Optimize Throughout the Marketing Lifecycle
Applying attribution before, after and during campaigns helps to inform upstream strategic thinking and enables downstream tactical optimization. This end-to-end analysis assists marketers in their efforts to drive improved effectiveness and efficiency for all elements of their media mix. Every insight along the way will help marketers optimize their campaigns, identifying which creative is resonating most, which audiences are most receptive and which channels and tactics are most effective.
۴. Build In Flexibility
To get the most from attribution clients must also rethink how they go to market. This line of questioning ensures that marketing programs have the built-in flexibility to activate recommendations that come out of real-time analytical insight. This may mean negotiating cross-platform deals that allow for budget shifts from less effective platforms to more effective ones, or better yet, deals that allow for wholesale budget cuts with the limited penalty. Companies should also consider holding back opportunistic ‘contingency’ funds that can be deployed to high-performing channels or tactics as they are uncovered.
So what’s next for attribution? As more user-level data becomes available for traditionally offline media channels, and as media buying continues its march toward a programmatic future, analytics will need to produce answers in an increasingly complex marketing ecosystem. Pressure will mount to integrate buying platforms with attribution methods to ensure that analyses are performed in real time. Additionally, the growing dominance of mobile as a media platform form is forcing marketers to create new processes to parse data and attribute credit.
Tomorrow’s winning marketers will be those who successfully create an integrated, holistic attribution capability that combines the high-level strategy of traditional, probabilistic attribution with the user-level, tactical insights of real-time data, and who work to activate the outcomes of the analyses in order to realize superior business value.
End/ Content Links