Are You Using the Return on Investment Metric Correctly?

Return on investment is a popular and potentially important metric allowing for the comparison of disparate investments.1 A critical requirement for calculating ROI is knowing the net profit generated by a specific investment decision. Most of the marketers we surveyed suggested that it was enough to know total profits and the investment to calculate ROI. However, this is incorrect; to calculate ROI accurately, you need to be able to estimate the fraction of profits attributable to the investment. As it is often hard to find the baseline — that is, what the profit would be if the investment had not been made — it can be difficult to calculate the incremental profit.

What’s more, ROI can be manipulated by cherry-picking the best projects: Being very selective might reduce total profits but increase the average ROI. In order to maximize ROI, you would invest only in the project with the highest return, even though maximizing net profit would require doing multiple projects. To illustrate, consider two potential investments of equal size. The first one has a ROI of 40%, the second a ROI of 30%. The second investment is still highly profitable — just not as attractive as the first. Making both investments would lead to a greater total profit, with an average ROI of 35%. By contrast, choosing only the higher-return project would mean lower total profits, despite its 40% ROI. As logical as this sounds, a large percentage of the marketing managers we surveyed incorrectly said that choosing a portfolio of the highest ROI investments was the same thing as choosing the highest total profits.

We would argue that the biggest challenge with ROI isn’t a technical deficiency but confusion over how it is used. In the 2014 CMO Survey,2 20% of respondents said they didn’t measure their marketing ROI. However, the responses from top marketers who said they did measure ROI were of greater concern. Fully one-fifth of the top marketers said they measured ROI using customer surveys, even though that’s not possible: Consumers don’t have the data (such as marketing investments and profits) to allow for ROI calculations. Another fifth of the respondents said they measured ROI using managerial judgment. This is also problematic because ROI is not subjective, but a metric with a specific definition.

Our survey also confirmed widespread lack of consensus over whether ROI is indeed a financial metric: More than half of the marketers we surveyed thought that ROI could be calculated using nonfinancial marketing data. When marketers resist using consistent definitions of measures such as ROI, it makes it more difficult to persuade nonmarketing executives that claims regarding marketing’s impact are credible.

Unmuddling ROI

Although ROI may not be a perfect metric, it is valuable to the extent that it can facilitate communication with nonmarketing colleagues.3 To communicate effectively, marketers must use terms in ways that nonmarketers can understand. Thus, a marketer should not use “ROI” to refer to every activity that has a successful outcome. A campaign to create awareness may have a positive ROI, for example, but marketers can’t prove this simply by pointing to higher levels of awareness. In order to calculate ROI, there must be a return (a profit associated with the investment) and an investment. Unless you have both, you cannot calculate ROI.

This material is excerpted from the article “The Metrics That Marketers Muddle,” by Neil T. Bendle and Charan K. Bagga. See the full article for advice on how to use five popular marketing metrics.