Meaningful Marketing Metrics

By Professor Byron Sharp, Director of the Ehrenberg-Bass Institute and author of Marketing: Theory, Evidence, Practice (second edition)

Technology and globalisation mean that many organisations are losing their location-based monopoly power and are having to become more customer-oriented. Consequently demand for marketing skills is rising.  Today we see banks, universities, even charities and hospitals with large marketing departments.

But technology also means that many basic marketing tasks will be automated. So the marketer of the future is going to need to have different skills. In particular they are going to be less of a ‘doer’ and more of a ‘thinker’. Analysis and evaluation will become a larger part of the job.

This isn’t to say that marketers will have to become statistical experts, but rather they will need to have a greater understanding of marketing metrics, and a greater ability to extract meaning from metrics. This means it’s especially important that marketing students learn these skills from the beginning of their studies in order to properly equip themselves as tomorrow’s marketing professionals.

Increasing amounts of money and management attention are being spent on marketing metrics. Marketers are demanding metrics for management purposes, and for reporting performance to company boards, shareholders and outside groups. (For example, companies now report metrics on their environmental record.) Research agencies, media companies and consultants are pumping out new metrics. Yet many of these metrics are misleading or unhelpful.

Marketing researcher John Bound once described his long career as a market research manager in large corporations as excelling in wasting his company’s money—with a note that he was sure that people do every bit as good a job today as he did then. Most market research is done technically very well, but it often fails to produce useful meaningful marketing metrics. The wrong things can be measured, in the wrong ways, and things may be misinterpreted.

There is no point in tracking and regularly reporting metrics that do not change, or do so very slowly. A good medical practice will measure the weight and blood pressure of patients each time they visit. But, sensibly, they will measure the height of adult patients only rarely. Conversely, many marketing metrics such as customer satisfaction and brand image are measured and reported far too regularly. Random sampling variation makes the figures wobble around a bit from survey to survey, providing the illusion of change. Much management time is then wasted coming up with erroneous explanations for movements that are simply random sampling variation. It keeps a lot of market researchers in business.

Table 1 shows evaluation of service performance scores across major US banks over a twenty-one-year period. Apart from a small dip around 1999 for all banks, performance seems to be around 70 to 80 out of 100. Much of that variability is likely to be accounted for by random sampling variation which creates the illusion of dramatic changes, changes that marketers would need to explain.

Table 1 Evaluation of service performance of major banks, 1995–2016

Table1_Banks

Source: American Consumer Satisfaction Index (2016)

Table 2 shows brand satisfaction with personal computers has not changed much over the past twenty-one years. Even the differences between the brands’ scores are consistent over time.

Similarly, there is little point in tracking and regularly reporting metrics that are already perfectly predictable from other metrics. Medical practitioners will not test male patients who complain of sickness in the morning for pregnancy—the fact that they are male ensures it is perfectly predictable that they are not pregnant. Many marketing metrics are predictable from market share or from how many customers the brand has (otherwise known as market penetration). The scores on these metrics are always higher for larger market share brands. Much market research, in effect, consists of asking respondents in numerous different ways how often they buy the brand.

Finally, there is little point in tracking and reporting metrics when you do not know what they mean, how they relate to other metrics and marketing actions, or what level they should be. Many special proprietary metrics sold by market research agencies fall into this category. Just because a measure is called ‘brand equity’ or ‘brand health’, doesn’t mean it necessarily connotes brand equity (the financial value of a brand’s market-based assets—mental and physical availability). This term is also sometimes used to refer to a consumer’s subjective assessment of how much better or preferred the brand is. Under this latter (not very useful) definition, higher quality, higher priced brands have higher equity.

Table 2 Brand satisfaction with personal computers, 1995–2016

Table2_PersonalComputers

Source: American Consumer Satisfaction Index (2016)

Imagine if your doctor measured your blood pressure and when you asked why and what it meant, she replied: ‘No idea … we were taught how to measure blood pressure at medical school, so I do it for all my patients—always have!’

Let’s continue with this medical example for a moment. When a doctor measures a patient’s blood pressure as, say, 140 over 90, they have a piece of data that is meaningless in its own right—is this high, low or just right? And why does this matter? Say the doctor takes the patient’s blood pressure a second time on another day using a different sphygmomanometer. The reading is still 140 over 90. The consistent results reassure the doctor about the reliability of their equipment and the credibility of the result. This is important, but the doctor still has nothing more than a meaningless (but reliable) piece of data.

To turn this data into information, the doctor needs scientific knowledge, such as how blood pressure scores relate to one another, how they vary between healthy and unhealthy individuals, and how blood pressure varies with age, gender and medication.

Given such knowledge, the doctor may then declare that 140 over 90 is high (not low or just right) and they know what this means for health.

Marketing professionals too need scientific knowledge in order to make their marketing metrics meaningful and to improve the quality of their management. Marketing research managers need such knowledge to avoid wasting money on pointless market research.

Professor Byron Sharp, Director of the Ehrenberg-Bass Institute, is the author of Marketing: Theory, Evidence Practice (second edition), How Brands Grow, and co-author of How Brands Grow: Part 2 with Jenni Romaniuk.

9780195590296

 

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s