How to predict the value of marketing activity

The way businesses go about marketing their products and services has undergone a major transformation in recent years. The global financial crisis placed additional pressure on marketing departments to demonstrate the worth of their promotional activities. At the same time, the expansion of social and digital media has broadened the potential scope and reach of marketing campaigns.

In response to these changes, the Chartered Institute of Management Accountants (CIMA) in partnership with the Chartered Institute of Marketing (CIM), the Direct Marketing Association (DMA) and Robert Shaw, the honorary professor of Marketing Metrics at Cass Business School and director of the Value Based Marketing Forum, have relaunched a report that examined the working practices of more than 100 organisations to assess how marketing creates value and to benchmark best practice.

The resulting report, Return on Ideas, suggests that optimum use of promotional budgets can be made when management accountants and marketing professionals combine their skills to make predictions about the effectiveness of marketing activities. These predictions enable them to focus on those that are most likely to generate value for the business.

The report recommends the use of the following techniques to help forecast whether an idea is worth investing in:

1. Look for analogies with existing situations. They can provide powerful evidence with which to evaluate a new idea. Where a substitute for a current product is proposed, a straight continuation of demand may be used to predict sales. This should take into account the length of the transition period as well as customer response to the product. Experimenting with a range of forecasting assumptions including best- and worst-case scenarios can be helpful.

2. Gain insight and foresight from customers. Good marketers consistently gather information about customer habits, preferences and intentions – all of which form the basis of subsequent predictions. Since habitual aspects of consumption are of major economic importance, observation of these habits can be a useful predictive method. For example, by tracking the eating habits of the general public, food manufacturers observed that convenience was becoming a priority for purchasers of salad vegetables. This insight led those manufacturers to launch successful bagged salad products.

Alternatively, customer insight can sometimes flag up instances when a particular approach would be unsuccessful because customers would be unlikely to break their existing habits.

Since the phrasing of a question can influence the respondent’s answer, customer analysis must be based on professional market research, with consumers categorised into segments according to their preference profiles. Rather than simply asking customers, “Would you buy this?”, the best practice involves asking them to compare paired products (or specifications or simulations of the product) and to assess their relative buying intentions. The accuracy of this “conjoint analysis” method is well-documented.

3. Conduct tests and experiments. Can the idea be mocked up for a simulation? For example, the release of a product in a particular town may be used to gauge customer acceptance and likely repeat purchase levels. As a control, the experiment should be conducted in at least one other location. The product should be sold in one of the towns with special packaging or a point-of-purchase display, and in the other without.

Simulated test marketing is an alternative, less expensive approach that combines testing with customer intention research. Online research surveys have recently grown in popularity because they can be set up quickly and the responses gathered swiftly.

4. Extrapolate from past experience. When ideas for the future are similar to the past, and data on the earlier ideas is available, extrapolations may be made. They can provide vital clues to the future, especially in markets where customer habits are important in shaping demand patterns.

Attention should be paid to the following aspects:

  • Examine and analyse revenue patterns into their component parts.
  • Determine any regular cycles, such as weekly and seasonal patterns.
  • Understand patterns of order sizes.
  • Examine correlations between prices and revenue patterns.
  • Investigate correlations between marketing activities and revenue patterns.

Employing two or more of these techniques in tandem helps prevent over-reliance on a single method. Complex scenarios may necessitate the services of an experienced statistician.

Of course, whichever method is used, it is important to acknowledge that prediction is never perfect and the inherent risks should be mitigated as far as possible.

Once a marketing campaign has been rolled out, the next step for management accountant and marketer is to demonstrate its value and assess the accuracy of their predictions. Good practice guidelines for linking marketing activity and expenditure to the financial benefits is detailed in the report.

Samantha White ( is a CGMA Magazine senior editor.