Will a celebrity endorsement be worthwhile and how much should we pay? Should we sponsor those golf tournaments again this year? How much should we budget for print advertising? These questions and many more keep marketing executives up at night. In the past, companies could test whether marketing activities were working or not. However today, with so many activities that affect marketing, executives are hard-pressed to identify the specific actions the drive marketing performance.
Enter econometrics, the science (and art) of statistical analysis to determine which combines of factors best explains a result. This analysis identifies activities that drive marketing performance and lift outcomes. The econometric analysis examines both sides of the equation. On one side are the inputs that include marketing activities, other controllable and noncontrollable variables, economic factors that might affect outcomes and competitor activities. On other side of the equation are the outcomes, which could include the number of new customers, retained customers, revenues, profits, customer equity and market share. All of the data is analyzed to determine relationships between the inputs and the outcomes over time. The results help marketing executives determine where to optimally allocate marketing resources to ensure the best outcomes.
For example, a company hired a famous television celebrity for radio ads that aired each week in several regional markets. They wanted to know if the celebrity added any lift to their sales. Because a lot of other marketing activities were occurring at the same time, it was difficult to isolate the celebrity’s influence. After an econometric analysis, the company learned that the celebrity had a substantial impact on sales in the markets where the radio ads aired. They decided to roll out the campaign nationally, and because they determined the exact impact, they were able to confidently renegotiate the contract with the celebrity.
As the cornerstone of econometric analysis, metrics play a critical role. Capturing appropriate data, both on the input and outcome sides, is vital to determining marketing accountability. The first step in the data identification process is to ask “What is the company’s goal?” As the outcome, the answer to this question dictates the type of metrics that will be used for the inputs. For example, because of the importance of customer lifetime value, many companies’ goal is to increase market share. For other companies, the goal may be profits. The inputs vary depending upon the goal or outcome that the company is seeking.
Once the goal/outcome is determined, the company focuses on the inputs to answer the question “What impacts the goal?” These metrics are segmented into several categories.
Marketing expenses are relatively straightforward. However, in the case of the radio ads, for example, the expenses are not listed when the invoice for the radio ad is paid. They are listed when the ad airs. In this way, expenses are always tied directly to a date when they affect the customer. Unfortunately, many companies track marketing expenses according to when invoices are paid, rather then when the impact occurs.
Other marketing activities are not tracked by expense. For example, the amount of money paid to execute a survey does not affect the outcomes as much as the survey value the consumer indicated on the survey. Surveys with high ratings may increase the number of new or retained customers as opposed to surveys with low ratings, which might have a reverse impact. Public-relations is another marketing activity where the amount of money spent may not affect the outcome. However, the amount of publicity generated in certain publications may have a substantial effect on the outcome. Publications can be rated by their circulation or eyeballs. And publicity can be rated as positive, neutral or negative.
Goals may be impacted by nonmarketing activities that are controlled by the company such as the number of stores or the number of salespeople. Goals are also impacted by noncontrollable variables, such as housing starts, seasonality and competition.
With lots of data resident in most companies, marketing executives have the capability to determine which marketing activities are driving performance. Armed with this knowledge, executives can more easily develop budgets, forecast sales and demonstrate marketing accountability.
Dan Otto is a principal at MarQuant Analytics, which helps companies define, capture, measure and analyze data to improve marketing productivity. He can be reached through the Web site at www.MarQuantAnalytics.com.
Barbara Lewis is a principal at MarQuant Analytics, which helps companies define, capture, measure and analyze data to improve marketing productivity. She can be reached through the Web site at www.MarQuantAnalytics.com.
