Stop Trying to Measure Marketing ROI

For the past several years, CEO's and CFO's have been demanding greater accountability from the marketing function, and they have been pressing marketers to prove the value of their activities and programs. In this environment, return on investment (ROI) has become the "gold standard" for measuring marketing performance. In a recent study by IBM, 63% of CMO's said that marketing ROI will be the most important measure of their success by 2015.

Return on investment has been a widely-used business performance metric for at least eighty years. It's a concept that business executives are comfortable with, so it's understandable that CEO's, CFO's, and marketers want to use ROI to evaluate the performance of marketing investments. In addition, many marketing writers and pundits contend that it's possible to calculate the ROI of almost any marketing activity. Unfortunately, however, it's not always easy or even possible to determine an accurate ROI for some marketing activities or programs.

Some time ago, I wrote a series of posts that discussed some of the issues relating to the use of ROI to measure marketing performance. You can find those posts here, here, here, and here. The point I want to make in this post is that if you can't accurately attribute revenues to a marketing activity, you can't calculate an accurate ROI for that activity.

The "revenue attribution" issue is particularly challenging for B2B companies that have complex, multi-step marketing and sales processes and long demand generation cycles. The diagram below illustrates the problem. This type of diagram is known as a cause-and-effect (or a "fishbone") diagram. Cause-and-effect diagrams are used in process improvement work to identify all of the possible causes of a given problem. In this case, I'm using the diagram to identify all of the marketing and sales interactions that occurred between a company and a prospect who ultimately made a $100,000 purchase.

In this hypothetical situation, the prospect was sent six lead generation offers and responded to the last of these offers. The prospect also:

• Received and responded to several lead nurturing offers
• Visited the company's website and viewed or downloaded several content resources
• Participated in several meetings and other activities with the company's sales rep

The question is:  How do you attribute the $100,000 in revenue to the marketing and sales activities that played some role in the purchase decision? What percentage of the revenue do you attribute to the lead generation programs, to the lead nurturing program, to the website, and to the direct selling activities? In reality, there's no way to accurately and reliably attribute revenue in these kinds of circumstances. Even our hypothetical prospect probably couldn't tell us how much each marketing/sales interaction influenced his/her purchase decision.

In these circumstances, it can be extremely difficult, if not impossible, to determine the ROI of an individual marketing activity. Many marketing activities do not produce revenues on their own. They must be combined with other activities to generate revenues. When a marketing activity is one of several interdependent activities that are all required to entice prospects to buy, you can measure the ROI of the whole group of activities, but not of any one of those activities.

This does not mean that you can't measure the performance of individual marketing activities. For example, you can and should measure the performance of your lead nurturing program by comparing the conversion rates of prospects who are nurtured with those who aren't. There are many metrics that can be used to evaluate the performance of marketing activities and programs, but you can't always use ROI.

OK, the title of this post may be a little misleading. ROI should be used in marketing whenever and wherever it's appropriate. Just don't try to use it everywhere.