Editor’s note: Mike Broom is CEO, Marketing Science, South Africa. This is an edited version of a post that was originally published by Bizcommunity under the title, “10 questions to ask the marketing director.”   

The economy and all companies within it have to continually make do with less credit, higher interest rates, wage increase expectations, higher input costs and increased pressure to be competitive. One way to think, and plan, around these pressures is to be more efficient. We look at marketing, in its broadest context, as namely increasing brand value and ROI.

The solution has been staring us in the face for the past 50 years, but the marketing community has preferred the fun of marketing, rather than the rigors of evidence-based decision-making processes, through the application of marketing science.

Consider the efficiencies and benefits that medical science has brought to mankind. It has developed to this stage over centuries with a dedication to evidence-based decision-making.

Think of your visit to a doctor. A few measurements – height; weight; conversation enquiries about general condition; urinary and bowel function; food and beverage intake; rounded off with a pin-prick – will enable the doctor to tell you if you are in danger of imminent death – a lingering, painful one – or give other advice that should be good news to you.

Nobody can give advice with such precision, accuracy and value in the field of marketing! Marketing is the only business activity where ignorance is tolerated.

Yet marketing science has been beavering away in academia and in pockets of excellence for years, understanding and growing marketing knowledge for decision support. The fundamentals have been in the public domain for decades. Proctor & Gamble, Unilever and BAT are among a handful of major corporations known to have established their own marketing science divisions. So what is your company doing about evidence-based decision support in marketing?


two paths representing decision-makingThe following are some of the key requirements of having necessary, up-to-date knowledge to be able to control the company’s marketing destiny with success.Achieving the level of marketing prediction common in medical science requires a synthesis and understanding of the interaction of measures and an advanced idea of modelling in decision-making, so that cogent alternatives of predicted outcomes can be the source of support for the decision. The following discussion is a personal perspective of 50 years of marketing research, including 20 years of studying and working with modelling in marketing.

1. How much profit comes from brands that did not exist in the company 10 years ago?

A company has to thrive to survive in a business environment where innovation is rampant. The rate of change is accelerating and the values and needs of children are no longer learned or molded by their parents. The question is posed as a measure of the extent to which your company is creating sustainable brand value. New products and brands are not successful if they are not profitable.

2. Does your company have a formalized ongoing new product development program?

Successful new products are rarely serendipitous insights. There needs to be a sustained process of careful investigation, research and testing. There are two useful ways of systematizing this: 

  • implement qualitative work to discover insights, dissatisfaction and gaps, which together with creative product engineering often results in successful products; and
  • estimate the market acceptance and forecast the potential sales of the concept.

There is a strong body of literature in the public domain about new product sales forecasting before launch. The one common theme is that the accuracy is extremely high. For example, when Indomie noodles were first launched in South Africa, the actual sales after 12 months was 5% higher than the forecast volumes estimated before launch. More importantly, when clients were advised not to launch and went ahead anyway, the product failed.

3. How many of your brand plans list increase awareness as a strategic goal?

In 1981 James Peckham’s “The Wheel of Marketing” was published. It brought to light the fact that ad spend itself does not create sales, it is ad share that does that, and even then, only with a delayed effect. Loosely, the difference between a brand’s market share and share of advertising is likely to result in market share changes in the next year in the same direction.Given that advertising is the most efficient way of creating awareness and that awareness is a necessary requirement for sales, the conclusion can be drawn that it is futile and expensive to increase awareness, because an increase in ad share pits your marginal increase against the entire category ad spend.

A more interesting route to market share growth would be to look at improving the trial rates, conversion rates and channels of distribution. The argument to increase awareness as a strategic goal is unachievable. An emphasis on trade action, couponing non-users and product improvement as a brand’s strategic goal are achievable.

4. How many new marketing campaigns have been implemented for the top five profit-earning brands over the past three years?

In general terms, the top profit earners are doing the job for you and a new marketing campaign may cause irreparable damage. The utmost caution should be exercised. There should be detailed econometric studies done on the advertising/market share coefficients.

In an example in the 1980s in South Africa, an underdog brand tried a number of different marketing campaigns to unseat the dominant brand. It was discovered through econometric studies that this spend was enhancing growth in the category and the dominant brand was growing as a result of those activities, not the brand spending the advertising. A textbook case of double jeopardy!

5. What is your company’s inflation-adjusted research budget compared to three years ago?

One would hope that it goes without saying that lower research funding results in lower quality information, and therefore lower quality decision-making. Of course, research budgets can be cut, but the quality of decision support should not be harmed.

One way of achieving this is by adopting a modelling approach in the management of brands. Often, it is worthwhile to go through the past three to 10 years of a brand’s history and assess the actions taken, the decisions made and the results on the life and profitability of the brand.

In my experience, when working on marketing science in a large beverage company, an analysis indicated that 80% of the promotions showed no positive effects, while 10% showed negative short-term effects. Of the remaining 10%, none showed a positive ROI.

6. Does your company have a policy of pretesting the strategic efficiency of advertising?

The question is what makes an advertisement effective – it is not just a question of whether the target market likes the advertisement, nor whether it provides relevant and interesting information. It should also deliver the strategic intent persuasively and believably.

7. Does your company have a formalized demand pricing modelling process in place?

Marketing departments should understand the price elasticity of their brands – one of the definitions of brand equity is the extent to which the brand volumes do not decline when competitors drop their prices. These are derived from studying sales data and also by conducting conjoint, Van Westendorp curve determination or Gabor-Granger studies, depending on their applicability in your primary category.

As an aside, it is particularly important to get this right for new product introductions. Dettol Hand Wash was the first on the market, and they started at a much higher price using modelling, rather than the cost-plus method of pricing. This enabled them to afford extensive advertising and have a war chest to fight off competitors.

8. Does your company use focus groups for decision support?

The standard reason given is that it’s quick, low cost and easy. But that doesn’t cut it when you can get a scientific evaluation of advertising alternatives, with diagnostics within three days from 200 people for less than the cost of two focus groups. Qualitative research is an indispensable tool in the marketing arsenal to gain an understanding of the complex and subtle product and brand relationships, but not for measurement.

9. Does your company run regular (at least once a year) brand health studies?

You really should, you know! I’ve come across a highly efficient brand health metric which gets 1,000 responses from a 10-minute survey on all brand health drivers. They are not that expensive and should deliver action points and be the source for all brand plans.

10. How many attributes per category, on average, does your company monitor?

Implicit in much marketing and advertising is the assumption of the Fishbein model, based on the attribute importance and brand association.

There’s nothing really wrong with it, and it does seem to make sense that people chose brands that they believe constitute the best attribute match for their needs.

While there may be elements of logical computation, in real life they involve a very small set of attributes, typically not more than six.

Wrong tool: big wrench with little nut

Making fewer mistakes

This article attempts to give some guidance on the role of successful brand innovation, development, planning and management and the role of marketing science and its role in assisting brands to be more profitable by making fewer mistakes.

A 10-point test of your company’s use of marketing science to create value in the business has been created. The link below will take you to the questionnaire and your score. Comments and suggestions on how to improve will be calculated and displayed as soon as you close the survey. Your identity will not be recorded, so any information provided will remain completely confidential.

Click here for questionnaire.