Recently a group of alumni from the University of Virginia Darden School of Business—some a few years out of school, some more—convened at 90octane to continue their education with professor emeritus Paul Farris. Farris is the co-author of such books as Cutting Edge Marketing Analytics: Real World Cases and Data Sets for Hands-On Learning and Marketing Metrics: The Definitive Guide to Measuring Marketing Performance. Although he’s recently retired, his academic work keeps him busy. At 90octane Farris shared ongoing research he and his colleagues are conducting to refresh their Article “Marketing Return on Investment: Seeking Clarity for Concept and Measurement”, published in the Applied Marketing Analytics Journal in April, 2015.

MROI is more relevant now than ever. Marketers work with a stunning array of channels, many of which come with advanced tracking capabilities. And with the corresponding growth in data at our disposal, we’re more accountable for the returns our spend is driving.

Calculating MROI boils down to this:

(Incremental financial value generated by marketing – cost of marketing)/cost of marketing

While the formula might be simple, applying it often isn’t. The methods of implementation can vary—resulting in substantial differences in the certainty of calculations.

Farris noted five different levels of measurement to which the MROI formula can be applied:

• Comparable costs: cost savings for achieving equivalently valuable customer contacts
• Funnel conversions (i.e., path to purchase): future period incremental sales and profits based on estimated conversion rates
• Baseline lift: current period incremental sales and profits
• Customer equity: changes in customer lifetime value
• Marketing assets: changes in brand and firm valuations (This spurred a discussion of what constitutes a marketing asset beyond brand, including customer file, customer data and distribution network. “Brand gets all the attention,” Farris said, “but these other things can be every bit as valuable.”)

And then, there are multiple lenses through which MROI can be viewed:

• Marginal returns (elasticity at the margin): What do we get from the last marketing dollar?
• Incremental returns: What do we get from marketing dollars over a certain period of time?
• Entirety (“the whole enchilada”): What do we get from the entire budget?

Not so simple after all, right? Acknowledging the complexity of the concepts he was sharing, Farris noted, “Our job as academics is to take things apart into so many pieces that you wonder if you can put it back together again.” Then he proceeded to do just that, by walking the audience through case studies corresponding to each level of MROI.

These examples highlighted the varying degrees of complexity—and accuracy—involved in calculating MROI. Take the customer equity case study: On the surface, the calculation seems straightforward.   

(Customer lifetime value (CLV) – acquisition cost)/CLV

But the devil in the detail is the retention cost buried in customer lifetime value. Some customers cost as much to retain annually as to acquire. Not to mention, in the disruption economy retention isn’t what it used to be. Switching costs are lower and competition is higher. Consequently, many marketers discount any ROI model that assumes retention beyond three years.

One of the alumni neatly summed up the complexity of MROI across all case studies: “None of these are controlled experiments.” That is, if you have multiple marketing different activities going on—as companies inevitably will—it’s hard to isolate just one.

And then there’s the question of just what metrics we should be prioritizing from an MROI standpoint. Impressions? Clicks? New customers? Referrals? Brand equity? They all vary in terms of our ability to measure and assign value to them. Further, given the ever-evolving nature of technologies and channels we have at our disposal, how far out can we make projections about any of these before they simply become a wild guess?

Which is not to say that efforts to calculate MROI are in vain.

“MROI is a big deal, and we should be using it,” Farris said. “But it’s easier to measure in some ways than others.”