An overwhelming amount of financial marketing executives—96 percent, according to The Financial Brand—feel that measuring ROI is a challenge. And 47 percent struggle to accurately quantify their department’s impact. That’s the bad news, the good news is that marketing is increasingly being seen as a revenue center within financial organizations—but with that comes a greater responsibility to deliver, measure and understand the ROI across the different touch points.
This becomes a greater challenge as both marketing and banking services become increasingly digitized and diversified across channels. For a regional bank or credit union, it is crucial to identify the best channels for reaching a desired demographic and then allocate marketing dollars effectively. As we move from viewing marketing as lead-generating to revenue-generating, it is ever more important to be able to attribute revenue to the appropriate efforts and channels.
Improving and assessing customer experience across media platforms is key for financial services, and a CMO now has so many options within the marketing budget, it is impossible to make impactful decisions about improving performance across channels without a robust attribution model. Banks need attribution models to help target customers with the right message, via the right channel at the right time.
Attribution Models—and Their Biases
There are several types of attribution models in use today. Some are more relevant than others, and each comes with an attribution bias toward certain channels and forms of media, which can result in less than optimal allocation of marketing dollars. Let’s take a quick look at these options:
- Direct attribution ties a specific customer behavior (opening a checking account) to a given marketing stimulus (such as a unique code or landing page). Although it is still relevant, direct attribution ignores the other marketing touch points that contributed to the behavior.
- Last touch/click attribution credits the customer behavior to the last known marketing touch or click. This is similar to direct attribution, but is typically used when there are no tracking codes linking customer behavior to a specific marketing stimulus. Here the marketer attributes the customer behavior to the last known touch—such as a Facebook ad for a mortgage program. Both direct attribution and last touch/click attribution ignore much of the actual buyer journey.
- A multi-full attribution model attempts to acknowledge the many touches that lead to a customer action, but goes about it in a less-than-ideal way. With a multi-full attribution model, credit for a customer action is attributed fully to multiple channels. The multi-full attribution approach is becoming more common, and that’s a problem (and one that is compounded by the silos that often spring up at financial institutions around marketers who cover different channels). When equal weight is given to all channels, there is no way to determine where marketing spending is being allocated effectively. Marketing channel silos also lead to data separation, with data being stored in fragmented pockets between various DMPs, vendors and partners contracted to capture it.
- Rules-based attribution is similar to the multi-full model described above, but in this instance marketers use a “common sense” approach to assign attribution across marketing channels. Where overlapping purchases are being attributed to multiple channels a rule is established that assigns a certain percentage of attribution to one channel and proportionally reduces the attribution by that percentage across other channels. Rules-based attribution has the right intent, but the problem is that guesswork and not analytics are being used to assign the percentages to each channel.
All of the attribution models listed above leave room for improvement. A good attribution model leverages analytics to derive insight from large pools of data.
Finding the Most Impactful Attribution Model
Which brings us to another model: fractional attribution. It uses math, science and statistics, not guesswork or “common sense,” to integrate data from across channels and turn that data into informed marketing decisions. Through proven analytic techniques, a weighting calculation is developed and applied to all the various touch points during a customer’s buying journey.
You want to know where you’re wasting marketing dollars? Fractional attribution is a critical tool for optimizing the performance of an incredibly fragmented and complex system of channels and media, both online and offline. No marketing science is perfect, but a robustly modeled attribution solution can help banks leverage marketing data and generate a higher return from their marketing spend—and outperform their competitors.
Stay tuned for our next blog post to find out how to transition to a fractional attribution model and start optimizing media and channel performance. Fractional attribution can help you break down the silos at your institution and tell you exactly where to move your money.
About the Author
As Chief Data and Analytics Officer for Harte Hanks, Korey Thurber manages ideation, development & delivery for customer analytics and data solutions. With more than 20 years of experience managing large international teams of data scientists, analysts & data solutions providers, Korey has helped many brands to successfully adapt & integrate analytics to develop, execute and optimize marketing strategies across diverse online and offline media. His extensive knowledge & experience has made him a valuable asset for many clients across multiple industry verticals including Financial Services, Wealth Management, Retail/CPG and Non-Profit.More Content by Korey Thurber