Version classiqueVersion mobile
OpenEdition Books

Changing Market Relationships in the Internet Age

 | 
Jean-Jacques Lambin

Chapter 5. Measuring operational marketing performance

Texte intégral

1The objective of this chapter is to raise the issue of marketing accountability. A central problem in business today is that marketing lacks the kind of accountability and metrics common to the rest of the corporation. For a very long time, this imprecision has been tolerated and has been excused because marketing was supposed to be inherently “creative”. Yet, as marketing consumes a larger and larger portion of the firm budget, the imperative grows to quantify marketing’s direct contribution to the bottom line.

5.1. THE CHALLENGE OF MARKETING MEASUREMENT

2Marketing is traditionally one of the least measured functions of the firm, despite the fact that it represents a significant share of companies’ budgets. This absence of marketing performance measurements is underscored by a study from the Chief Marketing Officer Council, representing 1 000 high technology companies like Adobe, AT&T, Cisco, Dell, Microsoft, Sony, etc., with combined annual revenues of $450 billion.

  • 1 CMO Council (2004), Measures and Metrics : The Marketing Performance Measurement Audit, Palo Alto, (...)

In a poll of 315 senior marketing executives, the CMO Council found that fewer than 20 percent of respondents have formal marketing performance measurement Systems in place. In addition, 80 percent are dissatisfied with their ability to demonstrate their marketing programmes business impact and value. Nearly 90 percent of respondents believe measuring marketing performance is a key priority for their organisation (CMO Council, 2004)1.

3These results are especially significant given the fact that technology companies tend to be among the heaviest spenders in marketing, with an average of more than 15 percent of revenues on marketing. This is several times the percentage spent on marketing by companies in many other major industries, including such marketing-driven industries as consumer packaged-goods.

  • 2 Palmer A. (2004), Introduction to Marketing, Oxford University Press, see chapter 12.

4Why so many firms are unwilling or unable to measure their marketing performance ? Different reasons are suggested by Palmer (2004)2.

  • Marketing is not perceived as a strategic issue by Board members having a weak market orientation.

  • Imagination, creativity and/or sheer determination are perceived as more important success factors than objective measurements.

  • Measuring marketing effectiveness is history ; fighting for the next battle should take the priority.

  • Marketing effectiveness cannot be reliably measured and is a too long and too costly process.

  • The environment changes too fast and results should be judged by the new realities not those prevailing at the time of the action.

  • 3 Eechambadi N., (2005). High Performance Marketing : Bringing Methods to the Madness of Marketing, D (...)
  • 4 USA Advertising Age Survey, 2003. See also Ruffle A. (2007), ROI : A Passing Fad or an Enduring Tre (...)

5Cultural resistance to measurement and analytic thinking can be problematic, but in the today’s business environment these arguments are not accepted any more, in particular because the functions of measurement go beyond explaining what has happened in the past. Measurements also move the firm forward towards appropriate actions and improvements. One cannot manage that which is not measured (Eechambadi, 2005)3. In the US, 70 per cent of marketers now say they use return on investment (ROI) calculations to guide long-term decisions on how they do business.4

• Development of appropriate marketing metrics

6While manufacturing and service organisations can quantify their costs down to a fraction of a euro and project their returns on investments, marketing remains a “dark science”, in which charismatic marketing practitioners can generate desirable results but cannot tell you how they achieved them.

  • 5 Churchill G.A. and Iacobucci D. (1992/2005), Marketing Research : Methodological Foundations, South (...)

7Marketing however has a long tradition of paying attention to measurement and the creation of metrics, as evidenced by the development of marketing research (Churchill and Iacobucci, 1992/2005)5. The problem is that most metrics used to assess the outcome of marketing activities are tactical and not directly relevant to the overall performance of the firm. Many companies do not even differentiate between necessary marketing expenditures (e.g. a catalogue) that are a cost of doing business, versus expenditures that are tactical with a short-term payoff (e.g. a promotion), versus expenditures that are investments in the future (e.g. a concept advertising campaign) with a longer-term payoff, such as building the brand’s capital of goodwill.

8Most marketers understand that while many marketing investments have some immediate payback, many require a bit of continuity to build their critical mass of contribution over time to achieve long-term objectives like building or maintaining brand equity or developing an emerging market. Unfortunately, accounting and finance people follow widely accepted accounting principles, requiring all marketing investments to be booked as expenses in the quarter in which they are incurred, even if the pay-off is expected for a much later period.

• The concept of brand equity

  • 6 Aaker D.A. (1991), Managing Brand Equity, New York, The Free Press.
  • 7 Nerlove M. and Arrow, K. (1960), Optimal Advertising Policy under Dynamic Conditions, Economica, 29 (...)
  • 8 Ambler T. (2003), Marketing and the Bottom Line, London, FT Prentice Hall, 2nd Edition.

9To refer to these lagged effects of marketing, the term brand equity as been popularised by Aaker in his 1991 bestseller book6, but the concept is much older and is due to Nerlove and Arrow (1960)7, who refer to the ‘capital of goodwill’ accumulated by a brand or a firm as a result of past advertising (marketing) efforts”. In accounting terms, brand equity is viewed as “... the intangible accumulated asset from past marketing efforts that has notyet been translated into profit”. This concept of an asset is nothing new in the field of accounting : “receivables” in the balance sheet stands for money that will be paid and “inventory” stands for goods that will be sold : Similarly, brand equity refers to “... the latent benefit of past investments before the sales emerge.” (Ambler, 2003, p. 51)8.

  • 9 Aaker D.A. (1991), op.cit. See also : Aaker D.A. (1996), Building Strong Brands, New York, The Free (...)
  • 10 Srivastava R.K. and Shocker A.D. (1991), Brand Equity : A Perspective on Its Meaning and Measuremen (...)
  • 11 Keller K.L. (1998), Strategic Brand Management, Upper Saddle River NJ : Prentice-Hall.
  • 12 Kapferer J.N. (2004), The New Strategic Brand Management, London, Kogan Page.
  • 13 Ambler T. (2003), op.cit.

10Other definitions of brand equity exist (Aaker, 1991)9, Srivastava and Shocker (1991)10, Keller (1998)11 and Kapferer (2004)12. The most compact one is due to Ambler (2003, p. 50)13 : “What is in people’s head about the brand”. It is interesting to note that, by “people”, one refers not only to consumers but also to the other market stakeholders, like distributors, influencers, suppliers, shareholders, employees, a view very much in line with the MO concept described in this book.

  • 14 Palda K.S. (1963), The Measurement of Cumulative Advertising Effects, Englewood Cliffs, New Jersey, (...)

11The existence of lagged effect of advertising - and by extension of operational marketing - is an established scientific fact since more than 50 years now, as the result of the seminal work of Palda (1963)14 and of many others. Marketers invest financial resources at some point of time and do not expect - except for response advertising and promotions - to gain the reward in sales in the short term, but rather to create, grow or maintain the brand’s capital of goodwill in people’s mind that will eventually translate into sales and profit. The idea was to find a concept summarising the strengths or “the health” of a brand. The brand equity concept was created because traditional data such as market share or volume sold were not satisfactory to reflect the value of a brand, since they are not taking into account the mental associations that existed in the customers’ mind. The concept has two faces. On the one hand, it gives a financial definition of the brand equity to evaluate the brand’s “financial” value (financial brand equity). It is especially important for financial analysts and companies to evaluate this strategic company asset. On the other hand, it covers the value of the brand from the customer viewpoint : the customer based brand equity, as a set of mental associations made by the customers and generating the brand’s strengths.

12Most CEOs today fully adhere to this idea, but they want more than an understanding of how marketing investments work, they want reliable measures of the impact on brand equity. In Table 6 are presented the marketing metrics most commonly used. These metrics are calculated differently in different sectors.

Table 6. Commonly used brand equity metrics

Table 6. Commonly used brand equity metrics

Source : Adaptedfrom Ambler (2003), Davis (2007) and Lenskold (2003)*

• The tyranny of measurement

  • 15 CMO Council (2004), op.cit.

13The difficulties of measuring performance vary with the type of marketing activities. According to the CMO Council’s report15, marketers say they are most capable of measuring direct mail campaigns, Web site and Internet search engine presence, telemarketing, and contact management programmes. They are least capable of measuring advertising and branding campaigns. A negative side effect of the measurement objective - known as the tyranny of measurement - is the shortermism it generates. As stated by

14Lapointe from MarketingNPV (2007)16, “... this increasing emphasis on quantifiable results has created an environment in which marketers have been politically pressured to shift an increasing portion of their program portfolio into tactics that pay back NOW or not at all.” (www.marketingnpv.com).

  • 17 Lodish L.M. and Mela C.F. (2007), If Brands are Built over Years, Why are they Managed over Quarter (...)

15The advent of store scanners, which gives managers real-time sales data revealing the immediate effect of price promotions, has contributed to popularise the short-term perspective. In addition, this orientation is exacerbated by financial analysts and rating agencies who focus on quarterly figures to value firms and advise clients. As put by Lodish and Mela (2007)17 “If brands are built over years ; why are they managed over quarters ?”

16How can the CMO choose between spending €1 million on a brand awareness advertising campaign or €1 million on a direct mail campaign ? High brand awareness affects every other marketing campaign and sales activity. It improves the returns from direct mail, from response advertising and makes the sales team more effective : But by how much ? Without clear measurable results, the impact is difficult to determine. That is why branding campaigns are considered more risky and therefore should have a higher return to account for these uncertainties. The temptation is to focus on marketing activities considered less risky and that can be measured well, while cutting back on longer-term investments in activities, not as measurable or measurable at a prohibitive cost, but that can build brand equity and augment shareholder value. This fact may explain the popularity of Internet advertising.

• Acquiring Customer and Market data

17Measuring marketing performance and productivity without data is of course a mission impossible. In addition to the traditional internal data recorded within the “order-shipping-billing” cycle, two types of external data are required : customer’s response data and competitive intelligence data.

18The customers’ response data are those related to the levels of the customer engagement cycle described below. These data are called external primary data, because they have to be generated and/or collected by the firm through market surveys or market tests or purchased to syndicated scanner data companies such as ACNielsen (www2.acnielsen.com), Research International (www.research-int.com), Taylor Nelson Sofres (www.tnsglobal. com) or Ipsos Mori (www.ipsos-mori.com). The formation of a customer panel or even a user community can also provide a platform for data collection.

19Competitive intelligence data are also required. Focusing exclusively on customers while ignoring competitor’s performance can be misleading. An 80 percent satisfaction level is great if the rate for competition is 70 percent, but not so good if theirs is 90 per cent. Thus, what matters is relative customer satisfaction, relative perceived quality, and relative price.

• The Customer Response Cycle

  • 18 Howard J.A. and Sheth J.N. (1969), The Theory of Buyer Behavior, New York, John Wiley and Sons.

20Market-driven management needs a marketing measurement framework to analyse the way marketing activities and investments influence customers in different ways at different levels of relationship development. The conceptual framework is provided by the response hierarchy model developed by Howard and Sheth (1969)18, which retraces the purchasing process followed by the customer and propose indicators of effectiveness for each level of the process.

21Table 7 : Key Measures of Customer Response

COGNITIVE RESPONSE

Awareness - Saliency - Familiarity - Recall - Recognition - Knowledge -Perceived Similarity.

AFFECTIVE RESPONSE

Attitude - Consideration -Affinity - Esteem - Relevance - Preference - Intention to Buy - Perceived value - Differentiation.

BEHAVIOURAL RESPONSE

Fact-finding behaviour - Trial purchase - Repeat purchase - Share of category requirement (exclusivity) - Loyalty - Bonding - Satisfaction/dissatisfaction.

Source : Lambin (2000)

22In this model, the various response levels of the customer can be classified into three categories : (a) cognitive response, which relates to retained information and knowledge ; (b) affective response, which concerns attitude and evaluation and (c) behavioural response, which refers to action, not only the act of purchasing, but also after-purchase behaviour. Table 7 describes the measures currently used by marketers for each response level. It is generally agreed that these three response levels follow a sequence and that the individual, like the organisation, reaches the three stages successively and in this order : cognitive (learn) - affective (feel) - behavioural (do). We then have a learning process which is observed in practice when the customer is heavily involved by his or her purchase decision, for example when the perceived risk or the brand sensitivity is high in the customers’ mind.

Table 8. Monitoring the Customer Engagement Cycle

Table 8. Monitoring the Customer Engagement Cycle

Source : Adaptedfrom Eechambadi N. (2005)

  • 19 Eechambadi N., (2005), op.cit. See Figure 4.3, page 86.

23An operational definition of the “learn-feel-do” model is the Customer Engagement Cycle (see Table 8)19 recognising different levels of customer engagement which lend themselves to measurement. Four sets of metrics can be developed to track (a) the marketing activities, (b) the customer impact, (c) the market impact, and (d) the financial impact of these actions, at each stage of the Customer Engagement Cycle.

  • Marketing activities : These metrics identify and quantify the particular actions undertaken to generate marketing results at each stage. Merely mapping out marketing activities, stage by stage, helps determining whether resources allocation is aligned with stated objectives.

  • Customer impact. This second set of metrics looks at outcomes or effectiveness indicators. It measures operational results like customer awareness, customer attitude and preference, customer intention to buy and satisfaction, etc. This information is useful for determining which marketing tools or programmes are effective in moving customers to the next stage or for reallocating marketing investments.

  • Market impact. This third set of indicators describes the resultant improvements in the brand or the firm’s market position, such as sales and market share, lower price sensitivity, higher customer retention, or stronger competitiveness. These improvements can be viewed as caused by the customer impact.

  • Financial impact. This fourth set of metrics is designed to translate operational marketing metrics into financial outcomes, incremental revenue, increased customer profitability, lower sales or service cost, stronger customer lifetime value, etc.

  • 20 Powell G.R. (2002), ROMI -Return on Marketing Investment, Albuquerque RPI Press.
  • 21 Ambler T. (2003), op.cit.
  • 22 Lenskold J.D. (2003), Marketing ROI, New York, McGraw-Hill.
  • 23 Davis J. (2007), Measuring Marketing : 103 Key Metrics Every Marketer Needs, Singapore, John Wiley (...)

24The financial benefits from marketing activities can then be evaluated in several ways ; return on investment (ROI) or on marketing expenditures (ROME), return on sales, return on assets, return on equity, net sales contributions. For a review of the methods used for measuring marketing productivity, see Powell (2002)20, Ambler (2003)21, Lenskold (2003)22 and in particular Davis (2007)23, who proposes 103 key marketing metrics.

5.2. IS MEASURING MARKETING ROI A REALISTIC OBJECTIVE ?

25By marketing ROI, we mean Return on Marketing Investments. For too long, marketers have not been held accountable for showing how marketing expenditures add to share-holder value. In today’s environment of financial austerity, Board members and Corporate Financial Officers (CFOs) are applying growing pressure on Corporate Marketing Officers (CMOs) to provide evidence that their marketing expenditures is worth it. And not only do marketers have to be accountable for their marketing expenditures, they also need to prove that they have the appropriate metrics and that measurement itself is feasible. In a growing number of companies, unless a given marketing investment can clearly be shown to generate sales, profits or leads, its value is questionable.

• Computing Marketing ROI

  • 24 Stewart D.W. (2006), Making Marketing Accountable, Graziado Business Report, Vol. 09, Issue 3.

26If marketing is to be a credible contributor to marketing success of the firm, it must speak the same financial language as the rest of the firm. And it must translate outcomes into economic metrics comprehensible outside the marketing department. Any marketing expenditure must be weighed against alternative non-marketing investments and measured against the potential for increasing profitability as a result of marketing in a given quarter versus not making the expenditure at all (Stewart, 2006)24.

  • 25 To define “return” some authors - for instance Powell (2002) - use the revenue generated by the mar (...)

27The computation of the ROI formula is straightforward25. The ROI is presented as a percentage calculated by dividing the return by the investment, where “return” is defined as : the revenue generated by the marketing investment, less cost of goods sold, less investment. The ROI value will be a positive percentage, a negative percentage, or zero. A positive ROI indicates that you have earned more than enough profits to cover your investment. If ROI is negative, you have lost money. Zero ROI is your income-based break even point.

28The ROI threshold, also called the hurdle rate, is the minimum ROI level for which a company will make investments. The ROI threshold is used to guide marketing managers in their decisions for which campaigns and which incremental investments should be pursued. It is the minimum ROI that management expects from any marketing investment. For example, if a company’s threshold is 25 per cent, funding will be provided for any investment opportunity that exceeds that level and rejected for an investment project below that level. A company may choose multiple ROI threshold based on the level of risk anticipated for the marketing investment. From a financial point of view, the ROI threshold should represent the company’s cost of securing capital.

• Computing the minimum return

29In computing marketing ROI, the difficulty lies in obtaining reliable estimates of the revenue generated by the marketing investment. While some forms of operational marketing investments, like direct mail, promotion or response advertising, are intended to generate immediate sales and lend themselves to measurement with a reasonable degree of accuracy, a brand image campaign or a sponsorship can create emotional connections and brand preferences that last over a very long period of time and may have implications on brand equity and on how the stock market values the company. For marketing investments of this type, it is difficult to force investments decisions into a standard marketing ROI equation, since the incremental revenue will be impossible to identify for each investment.

  • 26 The formula is derived from the ROI equation and is : Investment x (1 + ROI target) = Return Target

30The problem can be approached from another perspective. Using the ROI threshold or a target ROI, the minimum return that must be reached to recover the investment can be calculated and converted into a sales target. If the ROI target is 20 per cent, the return target formula is obtained by multiplying the investment by 1 plus the ROI target. For example, if the investment is €800 000, the minimum target return will be €960 000 (or €800 000 x 1.20).26 Alternatively, dividing the marketing investment by the gross profit margin (in value or in percent) will determine the sales (volume or value) required to recover the marketing investment. Without this type of analysis, sales target may be set too low and investments decisions can be made that do not recover the investment. In this approach, the marketing teams responsible for creating the brand image campaign are induced to recognise that they are expected to generate incremental revenue.

  • 27 Lenskold J.D. (2003), op.cit.

31Once the minimum target returns is determined, the CMO can convert this figure into sales objective to be achieved by the operational marketing campaigns having received the support of the brand image campaign or of the sponsorship. As explained by Lenskold (2003, p. 157)27, the CMO has three options : aggregate, allocate, or ignore.

  • Aggregate. Where a sponsorship or a brand image advertising campaign can be matched with operational marketing campaigns that generate sales through its support, the aggregated ROI is a strong measure.

  • Allocate. For a general brand image campaign that supports a number of customer segments, products, or markets, the target return can be allocated into the ROI calculations for the associated operational marketing campaigns. The allocation process is justified provided that the target audience, the products or services promoted the message and positioning and the timing of the campaigns are the same.

  • Ignore. Where brand image advertising has such a broad impact and longterm benefits, it is best to ignore this type of investment for ROI measurement and assess its effectiveness using brand equity metrics.

32The allocation process will provide insights into how operational marketing managers perceive the value of a sponsorship or brand image support. If the general consensus is that this type of general advertising is not providing a lift in sales, additional testing or analysis may be justified.

• Measuring the long-term effect of marketing

33Long-term effects of marketing are much harder to measure. A very large number of studies on the impact of marketing activities on brand performance are concentrated on activities having a short-term impact. A limited number of studies have analysed the longer-term impact of marketing instruments. As already underlined, one plausible reason many firms adopt a short-term emphasis on marketing activities is that they have a large short-term effect that lends itself to easy measurement.

  • 28 Ataman B., van Heerde H.J., and Mela C.F. (2006), The Long Term Effect of Marketing Strategy on Bra (...)
  • 29 The Principal-agent Theory has been initially developed by : Alchian A. and Demsetz H. (1972), Prod (...)

34Ataman, van Heerde and Mela (2006)28 propose another explanation referring implicitly to a “Principal-agent” problem (Alchian and Demsetz, 1972)29, where the interests of an agent (the brand manager) are not aligned with those of the principal (the shareholder). Brand managers have a brief tenure in which to be promoted, often spending a year before moving to another assignment. As such, long-term effects benefit their successor (or the shareholder), while short-term effects benefit them. Since there is little incentive to invest in long-term brand building, brand managers may choose to ignore the instruments that do lead to beneficial long-term effects, such as concept advertising, new product introductions or improved distribution that take months or years to manifest.

  • 30 eNumerys Global (2007, An Analytical Approach to Balancing Marketing and Branding ROI, a white pape (...)

35In a white paper published by the marketing research firm eNumerys Global (2007)30 www.enumrys(.com) it is suggested to leverage consumer-based equity metrics to measure the long-term effects of marketing on sales, in addition to the short-term effects measured by the marketing-mix models utilising a three stages approach depicted in Figure 11 (flow 3). This would provide a more complete measure of marketing ROI and bridge the gap that exists in how brands are measured and valued and the process of marketing budget allocation that eventually drives brand value.

Figure 11. Measuring short and long term marketing effectiveness

Figure 11. Measuring short and long term marketing effectiveness

Source : Adaptedfrom eNumerys Global White Paper (2007)

36In B2C markets, consumer tracking surveys regularly measure constructs like brand awareness and knowledge, brand interest and brand intention to buy. Similarly, more specific indicators of brand equity could be monitored over time. Here are four examples of indicators that could be introduced in a scale (Likert type) with value of (1) for strongly disagree and (7) for strongly agree.

I am ready to pay a higher price to buy brand A
If brand A is not available in my usual store, I am ready to visit anther store.
Brand A has element of uniqueness of value for me
Brand A is superior over its direct competitors.

37In addition to the functional benefits of the brand, attitudinal and emotional indicators reflecting aspirational values of consumers are also included. Examples could be :

Brand A makes me feel confident of my look
Brand A makes me feel like a great mom

./.../

38Monitoring the evolution of the functional and attitudinal indicators and relating their changes with the marketing activities undertaken to support the long term value of the brand equity, can be part of a more general evaluation system of the firm’s strategic brand management.

5.3. ARE MARKETING MODELS USEFUL AS DECISION SUPPORTS ?

  • 31 Bass F.M. and Buzzell R.D. et al. (1961), Mathematical Models and Methods in Marketing, Homewood Il (...)
  • 32 Montgomery D.B. and Urban G.L. (1969), Management Science in Marketing, Englewood Cliffs New Jersey (...)
  • 33 Little J.D.C. (1970), Models and Managers : the Concept of Decision Calculus, Management Science, 1 (...)
  • 34 Lilien G.L. and Rangaswamy A., (1998), Marketing Engineering, Computer-Assisted Marketing Analysis (...)
  • 35 Lilien G.L., Kotler P., and Moorthy (1992), Marketing Models, Prentice Hall International, Inc. NJ (...)
  • 36 Leeflang P.S.H., Wittink D.R., Wedel M., and Naert P.A.V. (2000), Building Models for Marketing Dec (...)
  • 37 Dyson P. (2002), Setting the Communication Budget, Admap, 433, 39-42.
  • 38 Ruffle A. (2007), ROI : a Passing Fad or an Enduring Trend ? Admap, Supplement February.

39Marketing models measure the impact of marketing activities, competitive effects and market environment factors on sales or market share of a brand. Mathematical model building has remained for years confined in the domain of Academia with the pioneering works of scholars like Bass and Buzzell (1961)31, Montgomery and Urban (1969)32, J.D.C. Little (1970 and 1979)33 who introduced the concept of decision support system (DSS), Gary Lilien and Rangaswamy (2004)34 and many others. For a comprehensive review of the marketing models proposed in the academic and professional literature, see Lilien, Kotler, and Sridhar Moorthy (1992)35 and Leeflang, et al., (2000)36. Building on these early works, many international marketing research companies have developed today marketing mix models useful to set communication budgets or to reengineer budget allocations among brands, regions or media, Mindshare (WPP), Data2Decisions, eNumerys Global being examples. For a review of the professional literature, see Dyson (2002)37 and Ruffle (2007)38.

40This methodology is increasingly used in the consumer packaged goods industry given the proliferation and accessibility of high frequency sales and marketing data on a weekly or monthly base. The technology has also improved. A brand manager today has the technological power on her/his desk to pull together all relevant sources of information, from hard sales data through consumer brand equity metrics and qualitative opinions about how the brand works. The pressure for greater marketing accountability already mentioned has also contributed to add impetus to this methodology.

• Connecting brand equity metrics and sales

41Brand equity metrics are currently collected by brand managers in the form of brand tracking surveys. The challenge is to uncover the connection between performance metrics such as awareness, preference, intention to buy, satisfaction to ROI. In other words can one translate a percentage point of awareness generated by a brand advertising campaign to a financial value of “X” euros ? The stand alone financial value of one point of brand awareness is zero, except if this metric is integrated within the context of the customer engagement funnel and viewed as a first step of the customer response chain leading to sale and market share. To build market share, the objectives are successively (a) brand awareness among the target segment, (b) a favourable attitude and interest in the brand, (c) intention to buy, (d) convenient brand availability and (e) satisfaction (see Table 5). Each step along the customer engagement chain indicates how customer response influences market share.

  • 39 Best R.J. (2004), Market-based Management, Upper Saddle River, Prentice-Hall, Third edition. This m (...)

42Best (2004)39 has suggested that market share can be estimated from the combination of a set of hierarchical marketing metrics indicating how customers respond to the firm’s marketing efforts. Using this model, the impact of a change of one marketing metric - for instance awareness - on the market share index can be calculated and converted into incremental revenue and in profit contribution after subtraction of the advertising cost. The calculated market share index is simply an approximation of what market share should be. The model provides a mechanism to assess the market share change when a certain level of improvement is made in a key marketing metric. It also enables the firm to identify sources of lost market share opportunity. The inputs of the model are branding equity metrics regularly collected by brand managers.

Marketing mix modelling

  • 40 Dyson P. (2002), op.cit.

43Marketing mix models are typically estimated using historical hard sales and marketing data on a weekly or monthly base using multivariate regression methods. The typical output is a decomposition of sales into baseline and marketing incremental, including sales due to each specific marketing activity. As illustrated in Figure 9 (flow 1), these models measure the short-term effect in sales due to individual marketing activity and can provide estimates of the ROI of each marketing instrument. Once these relationships establish, the model can be used to simulate the sales impact of changing investments in the different marketing instrument, to reallocate the marketing budget and to generate sales forecast and run “what if scenarios. As underlined by Dyson (2002)40, to be successful the modelling approach requires frequent interaction and communication between the modelling and marketing teams.

  • 41 Weaver K. (2004), Calculating the Payback from Advertising, Financial Marketing, August.

44Weaver (2004)41 from Data2Decisions Ltd reports that significant financial benefits result from using software systems that calculate pay-back from advertising. Sales can increase by around 10 per cent simply by re-phasing media spend. On a wider scale, advertising effectiveness can be improved by 25 per cent to 50 per cent by using the results of ad spend assessments to re-engineer budget allocations.

• Limitations of marketing-mix models

  • 42 Wittink D.R. (2005), Econometric Models for Marketing Decisions, Journal of Marketing Research, 42, (...)

45Marketing mix models have two important limitations. First, since they are based on an analytical assessment of the past, they will provide unreliable forecast when applied in situations where important changes are underway. For instance, when the Internet is transforming decision processes. When consumer decision processes, media, distribution channels are stable, these models work well. Thus, a marketing model is not a cure-all and blind faith in its results can be deceptive. As lucidly explained by Wittink (2005)42 “... many marketing models are, at best, descriptive. In that case, the estimated effects may properly describe how consumers act under prior conditions. As long as researchers do not capture how consumers may change of behaviour as a function of change in market conditions, the models will fail to make correct predictions of market place outcomes”. Some markets are easier to model than others. Fast moving consumer goods suit sales modelling analysis because they have short consumer purchase cycles. Also sales are tracked accurately and in detail using weekly data from scanners. The automotive sector is much harder to model because of its much longer purchase cycle.

  • 43 Lodish L.M. and Mela C.F. (2007), op.cit.

46There is a second important limitation. Standard marketing mix models only account for the short-term sales effect due to marketing activities and ignore the impact of marketing on brand equity (see Figure 9, flow 2). Thus, calculating marketing ROI on the base of the short-term effects only can be seriously misleading since the long-term effect of the different marketing instruments can be very different. For example, a brand image advertising campaign has generally a very modest positive short-term effect on sales but can have a much stronger long-term effect through its impact on brand equity metrics like awareness, esteem, or preference. According to Lodish and Mela (2007, p. 108)43, the long-term effect of advertising can be 60 per cent greater than its short-term impact. On the contrary, promotions or temporary price cuts generally have a strong impact on short-term sales, but their total impact is smaller than their short-term effects, due to stockpiling and competitive retaliation. In addition, they do not contribue to reinforce brand equity. In fact, frequent promotional discounting may even lower brand equity by commoditising the brand through increasing consumer focus on pricing.

• Who should be in charge of measuring marketing performance ?

  • 44 Delegge P. (2007), The Bottom line on Marketing Accountability, Marketing Today, December 17.
  • 45 Ambler T. (2000), Marketing Metrics, Business Strategy Review, 11, 2, 59-66.

47To avoid the pitfall of shortermism mentioned above, CMOs need to sit down with CFOs to determine the appropriate marketing measures and who is best suited to monitor these measures. Delegge (2007)44 considers that the Finance Department should take the responsibility for determining, managing and monitoring financial and non-financial metrics. Ambler (2000, p. 65)45 proposes the following reasons for this transfer of responsibility.

  • Marketers are widely seen as selective and/or manipulative in the way they present information. Independence would add credibility.

  • Metrics are not high on marketers’ priorities. Marketers are more interested in making runs than in scoring. Perhaps this is as it should be.

  • Marketing information is widely dispersed in large organisations. Only part of it exists in the marketing department.

48We do not agree with this very negative view of marketers even if we believe that the financial accountability of marketing should be improved. A partnership between the Finance and Marketing departments should facilitate cross functional coordination and thereby reinforcing marketing credibility.

Notes

1 CMO Council (2004), Measures and Metrics : The Marketing Performance Measurement Audit, Palo Alto, CA : CMO Council, April.

2 Palmer A. (2004), Introduction to Marketing, Oxford University Press, see chapter 12.

3 Eechambadi N., (2005). High Performance Marketing : Bringing Methods to the Madness of Marketing, Dearborn Trade Publishing., chapter 4, p. 64.

4 USA Advertising Age Survey, 2003. See also Ruffle A. (2007), ROI : A Passing Fad or an Enduring Trend ? Admap, Supplement February, 9th Edition.

5 Churchill G.A. and Iacobucci D. (1992/2005), Marketing Research : Methodological Foundations, South Western Thomson.

6 Aaker D.A. (1991), Managing Brand Equity, New York, The Free Press.

7 Nerlove M. and Arrow, K. (1960), Optimal Advertising Policy under Dynamic Conditions, Economica, 29, 131-145.

8 Ambler T. (2003), Marketing and the Bottom Line, London, FT Prentice Hall, 2nd Edition.

9 Aaker D.A. (1991), op.cit. See also : Aaker D.A. (1996), Building Strong Brands, New York, The Free Press.

10 Srivastava R.K. and Shocker A.D. (1991), Brand Equity : A Perspective on Its Meaning and Measurement, Cambridge, MA : Marketing Science Institute, working paper 91-124.

11 Keller K.L. (1998), Strategic Brand Management, Upper Saddle River NJ : Prentice-Hall.

12 Kapferer J.N. (2004), The New Strategic Brand Management, London, Kogan Page.

13 Ambler T. (2003), op.cit.

14 Palda K.S. (1963), The Measurement of Cumulative Advertising Effects, Englewood Cliffs, New Jersey, Prentice-Hall Inc.

15 CMO Council (2004), op.cit.

16 LaPointe P. (2007), Bridging the GAAP, Marketing NPV, 4, 1. (www.marketingnpv.com).

17 Lodish L.M. and Mela C.F. (2007), If Brands are Built over Years, Why are they Managed over Quarters ? Harvard Business Review, July-August, 104-112.

18 Howard J.A. and Sheth J.N. (1969), The Theory of Buyer Behavior, New York, John Wiley and Sons.

19 Eechambadi N., (2005), op.cit. See Figure 4.3, page 86.

20 Powell G.R. (2002), ROMI -Return on Marketing Investment, Albuquerque RPI Press.

21 Ambler T. (2003), op.cit.

22 Lenskold J.D. (2003), Marketing ROI, New York, McGraw-Hill.

23 Davis J. (2007), Measuring Marketing : 103 Key Metrics Every Marketer Needs, Singapore, John Wiley and Sons Asia.

24 Stewart D.W. (2006), Making Marketing Accountable, Graziado Business Report, Vol. 09, Issue 3.

25 To define “return” some authors - for instance Powell (2002) - use the revenue generated by the marketing investment in place of the gross margin. It is the amount that goes to the bottom line, not in the top line that should be used. The risk is to overestimate the profit contribution of marketing investments. For more details see Lenskold (2003), op.cit., chapter 6.

26 The formula is derived from the ROI equation and is : Investment x (1 + ROI target) = Return Target.

27 Lenskold J.D. (2003), op.cit.

28 Ataman B., van Heerde H.J., and Mela C.F. (2006), The Long Term Effect of Marketing Strategy on Brand Performance, Submittted to : Journal of Marketing Research, Second revision, July 2006.

29 The Principal-agent Theory has been initially developed by : Alchian A. and Demsetz H. (1972), Production, Information Costs and Economic Organization, American Economic Review, 62, 5, 777-795.

30 eNumerys Global (2007, An Analytical Approach to Balancing Marketing and Branding ROI, a white paper. (www.enumerys.com).

31 Bass F.M. and Buzzell R.D. et al. (1961), Mathematical Models and Methods in Marketing, Homewood Ill., R.D. Irwin Inc.

32 Montgomery D.B. and Urban G.L. (1969), Management Science in Marketing, Englewood Cliffs New Jersey, Prentice-Hall.

33 Little J.D.C. (1970), Models and Managers : the Concept of Decision Calculus, Management Science, 16, pp. 466-85. See also Little J.D.C. (1979), Decision Support for Marketing Managers, Journal of Marketing, 43, 9-26.

34 Lilien G.L. and Rangaswamy A., (1998), Marketing Engineering, Computer-Assisted Marketing Analysis and Planning, Reading Mass., Addison Wesley Longman, Inc..

35 Lilien G.L., Kotler P., and Moorthy (1992), Marketing Models, Prentice Hall International, Inc. NJ : Englewood Cliffs.

36 Leeflang P.S.H., Wittink D.R., Wedel M., and Naert P.A.V. (2000), Building Models for Marketing Decisions, International Series in Quantitative Marketing, Springer.

37 Dyson P. (2002), Setting the Communication Budget, Admap, 433, 39-42.

38 Ruffle A. (2007), ROI : a Passing Fad or an Enduring Trend ? Admap, Supplement February.

39 Best R.J. (2004), Market-based Management, Upper Saddle River, Prentice-Hall, Third edition. This model is mainly applicable in the field of fast moving consumer goods.

40 Dyson P. (2002), op.cit.

41 Weaver K. (2004), Calculating the Payback from Advertising, Financial Marketing, August.

42 Wittink D.R. (2005), Econometric Models for Marketing Decisions, Journal of Marketing Research, 42, February, 1-3.

43 Lodish L.M. and Mela C.F. (2007), op.cit.

44 Delegge P. (2007), The Bottom line on Marketing Accountability, Marketing Today, December 17.

45 Ambler T. (2000), Marketing Metrics, Business Strategy Review, 11, 2, 59-66.

Table des illustrations

Titre Table 6. Commonly used brand equity metrics
Légende Source : Adaptedfrom Ambler (2003), Davis (2007) and Lenskold (2003)*
URL http://books.openedition.org/pucl/docannexe/image/1651/img-1.jpg
Fichier image/jpeg, 404k
Titre Table 8. Monitoring the Customer Engagement Cycle
Légende Source : Adaptedfrom Eechambadi N. (2005)
URL http://books.openedition.org/pucl/docannexe/image/1651/img-2.jpg
Fichier image/jpeg, 504k
Titre Figure 11. Measuring short and long term marketing effectiveness
Légende Source : Adaptedfrom eNumerys Global White Paper (2007)
URL http://books.openedition.org/pucl/docannexe/image/1651/img-3.jpg
Fichier image/jpeg, 239k

© Presses universitaires de Louvain, 2008

Conditions d’utilisation : http://www.openedition.org/6540