Finding a measure that gives a fair assessment of a company's overall marketing effectiveness has so far stumped the best brains. What's needed is a system to identify the key indicators of performance and a standard language for what they are talking about. Alan Mitchell Mitchell.
What gets measured gets done. It's the maxim managers have taken to heart so completely that now, what gets done, is measurement. The marketers among them, however, appear to have lost their way in the measurement maze. Many measures of marketing effectiveness can be called upon. But which few should be chosen?
The differing measurement routes adopted by the following three disparate companies exemplify the range of approach available. Take SoftCo, for example. SoftCo (not its real name) is a company with an obsession. Some firms live and breathe total quality or innovation, but SoftCo is just as voracious when it comes to information. Every time a potential customer for this specialist engineering software company responds to a marketing initiative - whether it's a simple telephone inquiry or a reply paid card asking for more information - the name, the date, the nature of the inquiry, the company's response, and the initial trigger is recorded.
By analysing which communication channels work best, and how best to respond to different types of customer contacts, SoftCo marketers calculate they have managed to streamline their marketing budgets and improve effectiveness enough to boost the bottom line by 80%. They have found, says Gordon Swarz, a London Business School researcher, that 'management intuitions are often not justified by the data they have in hand. If they build a model they are often given significant surprises.' The business marketing company, 3M, has a different approach to measurement. 3M is a sprawling beast with products from Post-It Notes to industrial abrasives to warehousing control systems. It has recently embarked upon a multi-million pound global advertising campaign to raise awareness of its name, its record in innovation and its huge variety of products. Of course, the company has carefully tracked how its target audience (everybody from the self-employed who work from home to chief executives of large multinationals) has responded, and is pleased with the result. Awareness of 3M is up six or seven points, and 22% more people think it is 'innovatory'.
But it has not yet led to any identifiable increase in sales or profits. How can it justify spending that sort of money when it is producing merely ephemeral results? Alan Herbert, 3M's UK marketing communications manager, replies: 'We can't find a way of doing so. In the end it's a leap of faith - that if they know you and think you're a quality company they are more likely to buy from you.' Xerox takes a very different tack. Xerox is a process, benchmarking-fixated sort of company and it studies every marketing activity it undertakes, dismantling it detail by detail. 'Marketing is a process like any other business process, not an art form,' says Keith Bater, development manager for the process 'market to collection', which covers everything from determining what the company wants to do to actually delivering the product. 'We are most uncomfortable if we can't measure it.' To improve Xerox's performance in these areas, Bater has helped initiate a marketing benchmarking group through the Cranfield Institute of Management. 'Most marketing performance measures focus on outputs,' comments Martin Christopher, the Cranfield academic who has recruited companies such as Rover, ICL, SmithKline Beecham, the Royal Mail and Guinness to the project. 'But marketing processes are the glue that binds everything the company does, and the measures of performance, efficiency and effectiveness need to be process related.' It is difficult to imagine more disparate marketing activities or more different approaches to the vexed question of measuring marketing effectiveness. The three activities described - reputation building, lead generation, and promise delivery - are complementary. But finding a measure, or a set of measures, that give a fair assessment of a company's overall marketing effectiveness has so far stumped the best of brains. A wide variety of measures exist - market share, brand awareness and esteem, brand profit, brand value, customer satisfaction, customer retention, and so on - but there is no consensus as to which the best ones are, and no one has yet devised a model which pulls the various strands together into a convincing whole.
'Companies are doing various bits and pieces, but there is no systematic approach,' comments Stan Maklan, of the ForeFront business marketing consultancy. London Business School marketing accountability expert Patrick Barwise agrees: 'While accountants do some pretty weird and wonderful things, at least they have a consistent language.' Barwise dreams of an ideal world in which there are a dozen or so generally accepted measures and companies choose five or six that are most appropriate to them. But, he laments, 'there is no standard language for what marketers measure.' Part of the business marketer's problem is that since the year dot he's been overshadowed by the high profile world of fast moving consumers goods (FMCG). The FMCG marketers' mega-advertising budgets have helped create a myth that this alone is 'real' marketing. And because companies lavish 10% or more of their company's sales on marketing communications, it is here that marketing effectiveness has been most intensively researched.
Business marketers have different priorities to their FMCG peers. They target smaller audiences and for them relationship building is far more important than the glamour and mystique of advertising 'creativity'. What gets their customers going are the hard facts of price, specification, quality and service, and what grabs their attention is direct marketing and salesforce endeavour. Branding is different too. While FMCG companies have traditionally hidden their corporate persona behind a phalanx of separate stand-alone brands, in business the company is the brand and the corporate reputation is crucial.
A complex problem faces business marketers. On one hand, long experience of direct marketing and salesforce management means they are adept at number crunching effectiveness - at the level of short-term responses. On the other hand, the complexity of purchasing behaviours, sales channels and media has confounded broader attempts at measurement, as 3M's candid admission demonstrates. When 'real' effectiveness lies in the intangibles of customer relationships and the company's overall reputation, how do you measure it? And how do you identify the benefits relationship or reputation building bring?
No one knows yet. But they're working on it. While 3M's corporate branding exercise is a leap of faith, it's also busy measuring other aspects of its marketing. Adam Brand, 3M's manager of marketing information services for Europe, is struggling to assess and improve the effectiveness of its marketing function. 'It's quite difficult because we actually say that everybody's role is marketing. We don't want it separated,' he says.
He is running a pilot study designed to tease out key selling and marketing competencies, such as the gathering and use of marketing information, new product development processes, distribution and channel management, brand management and pricing. This involves identifying the questions each of the staff involved need to be able to answer. If it is pricing, for example, does he understand the range of pricing strategies available, is he capable of monitoring competitors' pricing strategies, is he aware of the legal and ethical issues, and so on? The aim: to track the development of that competency over time.
As part of its attempt to introduce a 'balanced scorecard' of business performance, 3M is also measuring things such as percentage of sales from new products, complaint rates to sales, and customer preferences alongside manufacturing and financial indicators. But there is a tendency for the number of measures to proliferate, and for the key indicators to get swamped. 3M currently has 13, and says Brand, 'we may well reduce that number'. Likewise, IBM tracks 24 different customer attitudes towards the company, its products and its brand personality, but now it's seeking to reduce that to just six.
IBM's means of measurement (see above) is predominantly quantative. However, companies such as Rover Cars, still favour the qualitative approach and focus particularly on customer attitudes. Rover's main marketing performance measures, says director of marketing strategy David Ruffell, are primarily external - customer (ie, fleet buyers, 'user choosers' and private individuals) perceptions of the company and its cars. On similar lines, Tim Ambler, Grand Metropolitan research fellow at London Business School, is seeking a measure of 'brand equity' based on customer perceptions of companies and their brands.
As business marketers start grappling with the issue of measurement and seek to apply what they have learned, the biggest benefit may be indirect: changing attitudes to marketing itself. Over-influenced by the example of consumer marketers, too many businesses have regarded marketing simply as a communications task. But now, the quest for marketing effectiveness is encouraging a much broader view. 'Marketing is the enabling process which aligns the internal capabilities of the whole company with the needs of the customer,' says Stephen Parkinson, professor of business strategy and director of Ulster Business School. 'Since marketing starts with the customer (need identification) and ends with the customer (need satisfaction) every action in the value-added chain between the two ends can be validly considered as a marketing process.' Parkinson is now in the second stage of a study for the Chartered Institute of Marketing of how a group of 44 chemicals, electrical, mechanical engineering, textiles, automotive and aerospace firms apply marketing processes and how this application - or lack of it - impacts on overall business performance.
The broader the definition of marketing, however, the more difficult measurement becomes. 'The big difference is between marketing and marketing communications,' says Peter Dart, who was involved in some of the earliest Unilever market modelling exercises and is now chairman of the Added Value marketing consultancy. 'If it's marketing communications, you can measure short-term promotional effects. But if it's marketing per se you have to take a holistic approach. Perhaps the simplest measure is profitability: business success.' Profit, however, is itself a notorious figure. And besides, it's got little predictive or diagnostic value: if measurement doesn't prompt learning and improvement, the measurement process itself isn't delivering value for money. The real challenge, suggests Rover's Ruffell, is to ask the questions that really tease out competitive advantages and problems. 'We try and thin out and understand the high leverage measures for us in the light of our particular strategic marketing objectives ... They may not be appropriate for any of our competitors.' So far, Barwise's hopes for a common language of marketing measurement look forlorn.
IBM's complex maths add up to a single number
One approach to marketing measurement, currently being piloted at IBM, seeks to come up with a single number which measures the value of the brand. The exceedingly complex technique, pioneered by Interbrand, works on the assumption that a brand's value is a function of two core factors: its earnings (a financial measure) and the security of those earnings (a marketing measure). The two scores are then combined to produce a brand value.
Determining the financial measure requires the company to estimate how much cash the brand generates over and above what it would generate if it were a non-branded product, and over and above what the funds would generate if they were sitting in the bank. Forecasts are then made for the next three or four years, and discounted back to present values.
To get the marketing measure, marketers are required to assess seven factors: the market, its stability, the brand's leadership position within it, market trends, marketing support, international presence, and its legal, trade mark and other protections. Each of these factors are weighted by marketers and each factor itself is made up of a number of weighted attributes. The leadership score, for example, may depend on the brand's market ranking, segment ranking, relative market share, relative price, segment share, and consumer attitudes to arrive at a final score of 'leadership'.
It's a time-consuming and expensive process: IBM has already spent 'a six figure sum' on its pilot study.
Originally the technique was applied to assessing the value of the IBM brand worldwide. Now it's being focused far more narrowly, to assess the contribution the IBM brand makes towards selling networks to UK businesses, for example. The problem, says networks marketing manager Kevin Bishop is that while the IBM name is crucial say, in mainframes, in networks, rivals such as Novell are better known. If the IBM brand is unimportant to networks purchasers, instead of trying to communicate direct to end users, could his marketing funds be better applied by promoting through resellers?
That's unlikely. But the fact that the question can now be investigated suggests that the whole brand valuation exercise is worthwhile. The absolute brand values themselves are relatively unimportant. It is the trends identified by a consistent, formal measurement process that matter.
Too much information and not enough guidance
The problems of measurement systems are legion. Even relatively simple indicators like response rates can be problematic. Counting the number of inquiries received from a direct mail shot or an advertising coupon is fine as a first step, but analysing how many of them turn into sales, and what sort of inquirer is likely to become a purchaser may be far more revealing - and difficult. Unless systems are set up for logging and tracking contacts with each potential customer, valuable insights into what works and what doesn't may be lost. 'It's hard work,' admits Keith Dunnell, MD of specialist business advertising firm Primary Contact. 'We have to bully many of our clients into doing it.' Getting the right information is crucial. Yet, according to Michael Brewer at consultancy firm Abram Hawkes, it's often not gathered. Marketers can invariably tell you how much they spend on advertising, brochures, trade fairs, and PR but they have no idea how this spending breaks down either for individual products or for different distribution channels. Once this information is unearthed damaging mismatches are often discovered. The company may be focusing advertising on its new range of products but the salesforce may still be concentrating its effort on the old.
A parallel weakness: not knowing your customer. 'Research into what the market thinks, feels and needs is the most underutilised tool in business to business marketing,' says Paul Hillson, of specialist agency IAS.
Companies can also overdose on information. Those wishing to track their performance often start monitoring a vast range of indices, from customer satisfaction through to the number of perfect deliveries (on time, in full, error free), through to purchaser attitudes (are they 'leading edge' friendly, and efficient?). Soon they're drowning in numbers. It's easy to end up with 'too much information and not enough guidance,' comments David Preston, marketing communications research manager at IBM.
Another mistake: measuring the wrong thing. Companies which track potential purchasers' attitudes may be wrong to assume that there is a correlation between attitudes and behaviour, for instance. Recent research suggests it's not always the case, warns Simon Knox of Cranfield School of Management. 'I suspect many marketers have got a spurious measure.' Knox is currently helping launch a Business Branding Forum which will attempt to tease out what role business brands have in influencing real buying decisions.
Spurious measures are not the only pitfall, however. So are irrelevant or unimportant ones.
'Marketers often create performance measures in non-threatening areas,' notes Kerry Turner, who has spent the last 10 years working on a Coopers and Lybrand system for marketing evaluation. They may, for example, spend a lot of time and energy trying to reduce the price they pay media owners for a certain targeted number of advertising 'opportunities to see'. But if advertising is the wrong approach in the first place, the savings generated are useless. And worse: if efforts are focused on improving performance in irrelevant areas it's a waste of precious management resource. 'If you don't measure what ought to be measured you get into all sorts of problems', says Turner. Useful measures, he adds, need to emerge from a carefully constructed model of how each particular market works. Timescales can be crucial. A sales chart may show dramatic sales increases following a special promotion or an advertising campaign. But if the tracking is extended, it may reveal that marketers have simply brought forward sales that would have been made anyway. Alternatively, a realistic analysis of the marketing investment may show real returns only coming through in three or four years while the corporate budget demands a two year pay-back - during which time scale the project fails to wash its face.
A related danger: inappropriate measures which send the wrong signals to staff. 'What gets measured gets performed,' warns LBS's Patrick Barwise. 'No matter what senior management's rhetoric is, if you bring in short term measures, you'll get short-termist behaviour.'