Are your campaigns minimising wasted creative and reducing your cost of sales? Good market research should be able to help you with your initial sale and subsequent sales. Sophisticated analysis can then assist you in segmenting your target groups and measuring your ROI.

The value of sales is not just the measurement of revenue, but profitability. You therefore need to take account of not only the initial sale, but the ‘lifetime value’ of the customer and the value of all referral sales too.

In this regard, the work of marketing writer Ray Kordupleski in the area of valuing customers may be of assistance. Kordupleski, author of Mastering Customer Value Brand Management, defines the CVA (Customer Value Added) rate as:

CVA = The average perceived worth of your company’s products and services
The average perceived worth of competitive offers

He says perceived worth is measured by asking customers, considering everything, “Are the products and services worth what you paid for them?” This question and the relative-to-competition CVA ratio are at the heart of the customer value management.

Setting your objectives

Objectives drive strategic choice. Strategy questions are ‘or’ questions: ‘You do this’ or ‘this’. Any ‘and’ question is tactical and isn’t related to ‘big’ objectives, but incremental gain. For example, an objective to attract Generation Ys drives strategy to target people under the age of 29 instead of over the age of 29. An objective related to how a direct mail execution should look is tactical.

In designing any direct marketing campaign you need to take account of immediacy objectives – the advertising and communication need to work quickly and be measurable. The best measurement is the immediate sales generated, and this is easily measurable in an absolute sense. Underneath all of this is the fact you are dealing with individuals, not brand categories.

Understanding those individuals, their lifestyle, family context and needs will ensure that any campaign design is well targeted. Market research is terrific in assisting with segmentation and needs analysis.

Don Pepper and Martha Rogers write about one-to-one marketing, where there is a need to know each customer as an individual and to design processes to meet individual needs, even when dealing with a large customer base.

In a similar vein, strategic database marketer Arthur Hughes says that surveys tell you more personal information about each customer than you could ever get from overlays or models. “Many people like to be asked their opinions or desires, particularly when they think that someone will pay some attention to what they say,” says Hughes.

Not all attributes you communicate will be equal. Some are worth far more to you than others. The trick is finding out which ones are most valuable, because a dollar spent against the most important areas is worth a lot more in sales than a dollar spent on less relevant ones.

Market research uses multi-variate analysis to identify the utility values of each ‘driver’. Prior to this stage, qualitative research can be used to ensure the most relevant attributes are covered in the research and the multi-variate analysis is meaningful.

As well as generating sales, all communication can help build long-term brand salience, and this too is measurable. Measuring the causal impact of the advertising and communication, however, presents more of a challenge.

One of the key benefits of DM is building a longer-term relationship and one of the best ways of conducting a dialogue with customers is the satisfaction survey. According to Hughes, “The information from satisfaction surveys should be stored in a database and used immediately as a part of an overall relationship-building strategy that could include:

  • making sure the customer purchases a replacement model at the appropriate time
  • making sure the customer comes back to you for spare parts, service or consumables, or
  • alerting the customer to other products you sell. 

“The strategy needs to be worked out in advance before the survey is even drawn up,” Hughes continues. “Each question should be designed with the strategy objective in mind. The appropriate response to the survey needs to be worked out and put in place and the cost of the survey should be built into a customer lifetime value model.”

The DM process

Let us look at a typical process path to generating successful sales.


Identify the target market:

  • demographically (age, gender, socio-economic status etc.)
  • geographically
  • geo-demographically
  • by needs
  • by psychographics
  • by social values
  • by product and brand usage
  • by existing attitudes to the product category and client brand, and
  • by current category and brand experience.

Concept testing

Having developed communication concepts against strategy, work out which of these concepts is likely to be the most effective in generating sales. Which of them best communicates against the ‘drivers’ most relevant to the target group? In testing or pricing alternative DM options, not only do you need to establish which is most effective in generating sales, but you also need to assess why. What lessons have you learned for the design of future marketing campaigns? Ultimately, the best test is parallel in-market testing of alternative options. The more you can pre-empt this with research, the faster you can implement the main program, and the more confident you can be of success.


Have you ever looked at modelling those people who respond positively to the offer and using this to pinpoint key opportunities for further DM, partially by CCD and location? Other disciplines use Choice Models and Decision Support Systems (DSSs) to model the value of different offers and the utility of the individual components of those offers. Using these tools, you can look at optimum pricing to generate maximum profits, by looking at the projected sales potential at each pricing level.


You can play ‘war games’ where, by reducing cost marginally, you identify incremental sales opportunity (assuming price sensitivity) in terms of:

  • the number of incremental sales
  • the value of these sales in absolute terms
  • the profile of those who come across, with a price decrease, and
  • where they come from.

The inverse is also possible, where you identify the impact of a price increase on sales (assuming price sensitivity) in terms of:

  • the number of lost sales
  • the value of lost sales and the resultant impact on profits, given a higher margin level
  • the profile of those who leave the brand, with a price increase, and
  • where they go to.

While this article is aimed at looking at ways to increase your ROI on direct marketing, I believe that intelligent use of research will itself provide an ROI on the efficacy of the DM campaign.

Effectiveness is measured by multiplying strategy by execution. For example, if you have a strategy value of nine out of 10 and an execution value of six out of 10, there is an ‘effectiveness value’ of 9 x 6 = 54.

If you only have a strategy value of four out 10 and execution value of seven out of 10, the ‘effectiveness value’ is 4 x 7 = 28 (half the previous example, even with an improved execution).

Getting your strategy right in what you communicate, where you communicate and to whom you communicate, will increase your success rating potential. Testing alternative executions of your DM against strategy within key target groups will ensure that you get the balance of the equation right.

Building an ongoing relationship

If you are a client, this is the least you deserve and should demand. As a supplier, your professionalism demands that you leverage all tools available to you. The fact that market researchers have not been beating a path to your door, nor been overwhelmed by briefs from you should not be a deterrent.

To conclude, I will borrow the wisdom of Arthur Hughes. According to Hughes, mistakes are often made through failure to understand the difference between market research and database marketing. “Modelling can help in deciding which individuals should receive promotions from the universe of those you could conceivably contact,” he says. “Using modelling, you can reduce the cost of your mailing and still maintain sales. Excessive reliance on modelling will divert marketers from their main objective: building an individual relationship with each customer, rather than spending money on appending external data (which may be inaccurate). It is better to use the same money to survey your customers and ask them why they bought your product, what they would like to see in the way of new products and services and what their plans are for the next year. Once you know these things, you can really build a relationship and your bottom line.”