Nike sprints ahead of the competition?

Malhotra and Birks: Marketing Research, 2nd Euro edition, Case studies

Case studies

Acknowledgements

We are grateful to the following for permission to reproduce copyright material:

‘Getting to know you’ (Acreman, S. and Pegram, B.), originally published in Research Magazine, November 1999, pp. 36–41.

In some instances we have been unable to trace the owners of copyright material, and we would appreciate any information that would enable us to do so.

CASE 1

Nike sprints ahead of the competition?

Nike was founded by Bill Bowerman, the legendary University of Oregon track and field coach, and Phil Knight, a University of Oregon business student and middle-distance runner under Bowerman. The partnership began in 1962 as Blue Ribbon Sports (BRS); their first-year sales totalled $8,000. In 1972 BRS changed its name to Nike, named after the Greek winged goddess of victory. Nike employs 22,000 people worldwide, from Nike World Headquarters in Oregon. With 1,500 employees working at the Laakdal Customer Service Centre, Belgium has the most employees of any EMEA (Europe, Middle East and Africa) country. The Netherlands is a close second, with 1,200 employees working at the European HQ in Hilversum. Nike is the number one athletic footwear company in the US and the number two American brand in terms of name recognition among overseas consumers, a status shared with IBM and second only to Coca-Cola. This high degree of recognition is probably one of the main reasons Nike has been so immensely successful. For the 2001 fiscal year sales in the US were $4,819m, up 2% from 2000. 51% of Nike revenue is generated by sales in the US. In Europe, sales were $2,585m, up 7% from 2000. Despite their healthy levels of growth, analysts forecast a change in Nike’s future. Although the company remains well managed, experts feel the company will have to work harder to grow in the future. Consumers are becoming less than enthralled with Nike’s $100-plus prices and are ‘clearing out’ their Nike-filled closets with other brand name clothing. Reebok and Adidas, the number two and three businesses in the industry, respectively, are hot on the heels of Nike. Adidas is giving Nike a run for its money in women’s sports, running and soccer. Reebok is focusing upon environmental issues in response to criticism of third-world exploitation levelled at the sportswear industry. Nike has adjusted its successful marketing formula, which was started in 1985. The company hopes to shift attention to its product innovation and how the company has evolved.

Perhaps such success should be attributed in part to its wildly successful concept-based advertising campaigns. The company uses a process that is often called ‘image transfer’. Nike ads almost never specifically place a product, or mention the brand name. A mood or atmosphere is created and then the brand is associated with that mood. ‘We don’t set out to make ads. The ultimate goal is to make a connection’, states Dan Weiden of one of Nike’s ad agencies. One ad featured the Beatles and clips of Nike athletes Michael Jordan and John McEnroe juxtaposed with pictures of non-celebrities also engaged in sports. It was used to infer that real athletes prefer Nike and that perhaps if the general audience buys the brand they will play better too. Nike’s unpredictable image-based ads have ranged from the shocking, such as its portrayal of real blood and guts in a ‘Search and Destroy’ campaign used during the 1996 Olympic games; to humorous, such as the first ad used to launch Michael Jordan’s Jordan brand wear. The latter advertising made the tongue-in-cheek suggestion that Jordan himself had a hand in production by slipping away from a Bulls’ game at half time to run over to his company and then making his way back in time for the game’s second half.

Of late, Nike has shifted to the new phase in its marketing campaign, emphasising more of its product innovation skills than the jokey, edgy attitude that it displayed in previous years. ‘We recognise that our advertisements need to tell consumers that we’re about product innovation and not just athletes and exposure. We need to prove to consumers that we’re not just slapping a whoosh [the company trade-mark] on stuff to make a buck’, says Chris Zimmerman, director of Nike’s US advertising. With the launch of the ‘I can’ campaign, Nike is showing less of the celebrity athletes that previously adorned its marketing output and showing more product usage than in the previous ‘Just Do It’ campaign. Competitors Reebok and Adidas have recently featured more product-focused ads and have met with a great deal of success. Despite this rearranged focus, Nike is not backing away from innovative marketing.

Also at the heart of Nike’s new strategy is the international arena. That may prove to be the most difficult element for Nike to undertake. According to one source there seems to be pretty strong recognition that in a few years Nike will be larger outside the United States than inside. The problem that the company faces is that even though international sales currently make up one-third of Nike’s business, international business still pales in comparison with the domestic sales. Nike would like to expand into the soccer and international sports arena, but to do so it will have to refocus marketing and distribution in order to re-establish itself as an authentic, technically superior sports shoe. In the future, Nike would like to build up its presence in the key markets of China, Germany, Mexico and Japan. Nike will focus its advertising on sports, and will feature sports that are of particular interest in a specific region. Nike realises that while it is ahead of the competition, it still has a long, long way to run.

References

Jeff Jensen, ‘Reebok and Nike go head to head in soccer battle’, Advertising Age 68(18) (5 May 1997): 4.

Jeff Jensen, ‘Product innovation steps into spotlight at Nike’, Advertising Age 68(36) (8 September 1997): 3, 48.

Isabelle Sender, ‘Just do it, abroad’, Chain Store Age 73(10) (October 1997): 121–4.

Questions

What is the management decision problem facing Nike as it attempts to retain its leadership position?
Define the marketing research problem facing Nike, given the management decision problem you have identified.
Develop two suitable research questions and formulate two hypotheses for each.
How can the Internet be used to help Nike in conducting marketing research?

CASE 2

Getting to know you

Markets are becoming ever more competitive, with new brands competing for a share of consumers’ hearts and minds. Building a relationship with consumers is a challenge facing all organisations, but particularly so in the case of ‘emergent drinkers’ – those of legal drinking age up to 25. These consumers are highly experimental, and our only safe assumption is that their consumption habits will probably not follow on from those of preceding generations.

In 1997 Allied Domecq Spirits and Wines (ADSW), , recognised the danger of being distanced from this crucial group, particularly across geographical markets. They were not looking to understand a current user group per se, but rather to gain insight into the factors influencing brand adoption as these young drinkers mature.

Working with Pegram Walters International (PWI), , a unique programme of research was created. The objectives went far beyond an exploration of the current usage and attitudes of these consumers towards spirits, but also encompassed an exploration of their personal values, their feelings about their lives, their universe, their hopes and dreams. The project required a willingness to think beyond current market conditions and business objectives.

The broad objectives of the research, covering an elusive respondent set, clearly required an approach that would be both informal and unconventional. It needed to venture beyond ‘traditional’ marketing research in order to maximise the quality of data. Moreover, because of the innovative nature of the project, it required a high level of openness, communication and trust between client and agency to ensure that the information was both usable and relevant. In research terms there were two clear challenges:

Gathering information from this difficult-access consumer group
Integrating the information back into the organisation.

Gaining access to the adult emergent drinker

It was believed that to gain a real insight into the emergent drinker community, two realities would have to be taken into consideration:

Nobody can understand a community better than the community itself.
Information alone cannot provide valuable insight. Insight can only be developed from the blending of community understanding with external analysis.

Access to the community was provided via the development of the ‘information gatherers’ (IGs) concept. IGs would be representatives of the adult emergent drinker target group. They would participate in the research in order to interpret the dynamics of their own community for us. To accomplish this, adult emergent spirits drinkers were recruited, who were required to understand the objectives of the research project and to be able to communicate concepts. In this way, they would not only provide feedback on their own needs and actions but, more importantly, they would also be able to gather and interpret information from their peer group. IGs would become effectively both respondents and researchers, with the ability to provide rich, value-added insight.

It was felt that one of the key successes of this programme was a policy of maintaining honesty at all stages of the programme. By being completely open about the objectives with participants, by sharing the hopes and expectations with them, the IGs were empowered to have a stake in the project. As a result, they felt as committed to gaining valuable insight as the client and research agency. However, the recruitment of IGs also required both the client and the research agency to step away from established comfort zones and let go of control – two key ingredients to any programme of innovation. Overall there were three stages to the research design in each market.

In the first instance one-hour depth interviews were conducted. There were three clear objectives for this stage of research, to:

Understand personal viewpoints on marketing and lifestyle issues;
Clarify and/or narrow down topics for subsequent exploration at the workshop stage;
Recruit appropriate ‘information gatherers’ (IGs).

Depth interviews were conducted to understand what was happening in respondents’ lives. They were invited to undertake ‘homework’ such as essays on their home lives, and to bring along items of personal importance to stimulate discussion. From this stage hypotheses were formulated on issues such as how they saw themselves and their future, relationships, self-discovery, and opting in or opting out of the system. In each market, from 20 depth interviews, 10 respondents were retained as IGs to accompany the researchers through the rest of the programme. It was believed to be important to conduct the bulk of the research in the environment in which alcohol was consumed. Leading-edge bars were rented out where 50 adult emergent drinkers were invited to participate in workshops.

From the time the participants entered the venue, the role of client and the research agency became purely observational, with the IGs leading the discussion throughout. A task guideline was designed which empowered the IGs with an understanding of the research needs, and then they were left to do it. As an additional record, the workshops were video-recorded. Because of the way in which they had been recruited, the IGs felt a real responsibility to get the right information. The participants felt comfortable within their peer group and, in the more natural bar environment, fed back real, relevant and honest information. Moreover, both respondents and IGs respected the process, allowing them to ‘buy into’ the research. On the night following the workshops, focus groups were reconvened with the IGs to discuss what actually happened, and their interpretation of what it actually meant. In this way, concentrated data were collected: what was observed, what consumers said, how it was reported back, and an initial understanding of what it all might actually mean.

Communicating the findings with the client: Allied Domecq Spirits & Wines

This is a preview of the whole essay

Communicating the findings with the client: Allied Domecq Spirits & Wines

To infuse the exercise with knowledge, learning and a sense of adventure, ADSW business teams were invited to spend a day of discovery. The day began by holding breakout sessions that included ADSW marketing and sales personnel, and their key agencies. The purpose was to gauge current assumptions about adult emergent drinkers, and where necessary to dispel some myths. They then ‘met’ the generation. It was felt that the best way to do this was to create fictional characters for the adult emergent drinker generation. These would enable ADSW marketing managers to visualise the consumers when developing new product development or communication strategies. The personalities created were brought to life using actors from the generation. In France, for example, the clients were able to meet Matthias, Stephanie, Seb, Justine and Stan.

These five characters symbolised the richness and diversity of the generation. They were not meant to represent a segmentation of the market – rather, they were intended to reflect a collage of adult emergent drinkers in order to help business managers enter into a relationship with this consumer group.

Each of the characters engaged with the audience via dialogue, discussing for example their lifestyle, behaviours, in/outs, values, concerns and expectations for the future, as well as their current attitudes towards alcohol. In addition, the audience was presented with workshop ‘souvenirs’, notebooks with pictures and ‘biographies’ of the character types where they could take notes during the presentation. The work groups were then reconvened to summarise learning. The effect was immediate: with the bar as a cue, business managers were able to step into a new world and easily meet and interact with their consumers. Moreover, their ‘consumers’ were eager to explain what was and wasn’t important to them. This multi-media/multi-layered presentation of findings allowed information to be assimilated both visually, audibly and kinaesthetically.

In order to ensure that the information remained topical, useful and easily accessible, it was felt important to create a vehicle for on-going communication and dialogue with the audience. To achieve this, a high impact ‘magazine’ was created to bring the research to life after the presentation. This is referred to as a magazine and not a research report, to reflect the lifestyle of the consumer group in question: it contained images, layouts and fonts typically associated with the generation. This magazine, together with the videos containing live footage of the actors and the workshop, was distributed throughout Allied Domecq Spirits & Wines.

On-going innovation

This research was considered to be an important exercise in terms of combining creativity of process and reportage with real business needs. It is often stated that researchers need to ‘get into consumers’ minds’, and to use creative/projective techniques to really understand what consumers are thinking. However, researchers often fall short in that it is frequently forgotten to devote the same amount of time to understanding and to the context of a clients’ businesses.

Such was the success of the research format that the research agency developed CommunityInsight, an information gathering process that aims to access primary target groups. CommunityInsight is based on the same two very simple premises:

Nobody can understand a community better than the community itself.
Information alone cannot provide valuable insight, which can only be developed from the blending of community understanding with external analysis.

Questions

What is the meaning of ‘access’ and why is it such a big problem for marketing researchers?
What are the advantages and disadvantages of using ‘Information Gatherers’?
What is a ‘leading-edge bar’ and what role did such a context for interviews play in this project?
What do you see as being the advantages and disadvantages of using actors to portray consumers as part of the oral presentation of research findings?
What is meant by ‘Nobody can understand a community better than the community itself’?
What is meant by ‘Information alone cannot provide valuable insight, which can only be developed from the blending of community understanding with external analysis’?

CASE 3

Danger! Celebrity in use

The theory behind the use of celebrities is that featuring stars in advertising has special cultural significance, borne from the unique way that they have constructed an image through various forms of media in the culture. In associating the star with the product, these special meanings are passed on to the products or brands. Celebrities have particular configurations of meanings that cannot be found elsewhere.

Celebrities are believed to possess dynamic qualities – attraction, sexiness and likeability, which can be transferred to products. For example, the Peugeot 406 acquired some of its ‘attractive and sexy’ attributes from Kim Basinger's appearance in its advertisements. Because they are famous, celebrities can attract and maintain attention by their presence in ads, and also achieve high recall results. Even though research findings are equivocal about the ability of celebrities to stimulate actual purchase behaviour, the positive impact of celebrity endorsers is well documented. When Michael Jordan returned to the NBA for his old team the Chicago Bulls in 1995, his return was calculated to be worth around $1 billion on the combined market values of the five companies he endorsed. Even though Pepsico has had bad experiences with celebrities, Michael Jackson and the Spice Girls were attributed with gains in share of 8% and 2% respectively in 1984 and 1997.

Celebrities can easily affect company or product image makeovers, as well as reposition an old brand or introduce a new brand. An example is Lucozade, which was able to achieve a new athletic and energetic image through association with well-known sports stars such as Daley Thompson, John Barnes and Linford Christie.

Celebrities with worldwide popularity can help global marketing communications, especially in advertising. Though Michael Jackson eventually turned into a disaster for Pepsi, the first year of the contract with Jackson saw an 8% gain in sales in a market where 1% is equivalent to millions of dollars. Some global celebrities choose to endorse products only in other countries to protect their image at home. For example, Mel Gibson and Woody Allen endorse products only in Japan. Pizza Hut increased market share by using global celebrities such as supermodels Cindy Crawford and Linda Evangelista, and Baywatch star Pamela Anderson for pan-European campaigns.

However, some companies have been embarrassed when their celebrity has become embroiled in scandal or controversy, for instance, Hertz (O.J. Simpson) and Pepsico (Michael Jackson). Most companies are smart enough to include clauses in their celebrity contracts for termination due to moral turpitude, and can purchase ‘death, disablement and disgrace’ insurance. They may still lose out on their investment and damage their image. A celebrity may disappear from the limelight before the end of a contractual term. This can be avoided, to an extent, by careful examination of the life-cycle stage of the celebrity: they should, ideally, be signed during the growth stage of their career, and terminated prior to decline – but this is not always easy to achieve.

A celebrity may become spokesperson for many products and become over-exposed. The spread of links over too many brands is likely to dilute the impact for any one of them. Possible examples are the Spice Girls (Walkers Crisps, Pepsi, Chupa Chups). Celebrities can be restricted by contract from working for competitors, or even any other brand, but this can be very expensive. It is unusual for a celebrity's image to change suddenly, but this can be a total disaster for a campaign. For example, French footballer David Ginola endorses L'Oréal's Elvive shampoo both because he is popular and because of his ‘clean, shiny and controllable’ hair. If he suddenly decided to shave his hair off, the campaign would be in ruins. A real-life example is Yardley's experience with Helena Bonham-Carter, who admitted in her first brand interview that she rarely used make-up. It can be necessary to explain very thoroughly to a celebrity exactly what their role and responsibilities for the brand are.

One researcher, Michael Kamins, employed marketing research techniques in order to explore the uses of celebrities in advertising more scientifically. Kamins states that three processes of social influence determine whether an individual will adopt the attitude an advertiser is trying to convey: compliance, identification and internalisation. Although the first of these factors is not relevant to Kamins’s study, the last two hold considerable implications for celebrity advertising. Identification, whereby individuals try to imitate another person because they want to be like that person, is the most important factor determining a celebrity’s influence in an advertisement. Internalisation occurs when individuals imitate another because they perceive the other person to be sincere and to have values similar to their own.

Kamins inferred that if both identification and internalisation could be achieved, the effectiveness of advertising would be increased. Therefore, he studied whether celebrities could increase the effectiveness of advertising through the identification component, and whether so-called truth in advertising (operationalised as two-sided advertising, or advertising that included both positive and negative aspects about a product) could increase effectiveness through internalisation. Furthermore, he wondered whether combining these two approaches resulted in even greater effectiveness.

In order to research this, a 2 × 2 factorial design was adopted. Sidedness (one-sided versus two-sided) and type of spokesperson (celebrity versus non-celebrity) were the two factors. Seventy-seven executives enrolled in an executive MBA programme were randomly assigned to four groups: one-sided/non-celebrity, one-sided/celebrity, two-sided/non-celebrity and two-sided/celebrity. Four advertisements corresponding to these criteria were made up, and each member of each group evaluated the appropriate advertisement on the basis of four variables:

Expectancy-value brand attitude (A).
Global brand attitude (B).
Global attitude towards the advertisement (C).
Purchase intention (D).

Expectancy-value brand attitude represented the degree to which the respondent believed the product possessed an attribute the advertisement claimed it had. Global brand attitude was a measure of how appealing the respondents found the product in the advertisement to be. Global attitude towards the advertisement was an evaluation of the advertisement’s effectiveness. Purchase intention indicated how likely a respondent was to purchase the product when an opportunity to do so came about.

Table 1 shows the mean (x) and standard deviation (sd) values, along with the number of respondents (n), for each variable across each of the groups in Kamins’s study. Note that the results from related groups can be combined to yield information on each of the four group characteristics (one-sided, two-sided, non-celebrity and celebrity) separately. Table 2 contains the ANOVA results for the effect of the independent variables of sidedness (E) and type of spokesperson (S). These results provide valuable information about the effectiveness of celebrity spokespersons in advertisements.

Although the above research results are useful, they are also very specific. Celebrity advertising can be researched in other ways. For example, dead celebrities have been shown to be hip, hot and safe. Although they are not cheap (advertisers do have to pay licensing fees to the celebrities’ estates), they are safe, because they can’t do anything unpredictable that might jeopardise a product’s image or embarrass the sponsor. Abbott and Costello have been used to sell bran cereal, Humphrey Bogart to lend flair to cellular phones and Diet Coke, and Charlie Chaplin to push IBM personal computers. This just goes to show that while the celebrities themselves may be dead, their use as advertising spokespersons has a long life.

TABLE 1

Means, standard deviations and number of subjects

Variable

Experimental Expectancy-value Global brand Global attitude Purchase

condition brand attitude attitude towards the ad intention

(A) (B) (C) (D)

(x) 7.97 3.47 3.4 2.22 (sd) 3.92 1.47 1.52 1.4

One-sided (n) 38 40 40 40

(x) 8.33 4.22 3.65 2.92

(sd) 5.32 1.6 1.62 1.44

Two-sided (n) 36 37 37 37

(x) 8.04 3.5 3.65 2.55

(sd) 4.73 1.55 1.46 1.38

Non-celebrity (n) 38 40 40 40

(x) 8.26 4.19 3.38 2.57

(sd) 4.58 1.52 1.67 1.56

Celebrity (n) 36 37 37 37

(x) 7.89 3.45 3.55 2.4

(sd) 4.48 1.57 1.39 1.5

One-sided non-celebrity (n) 19 20 20 20

(x) 8.04 3.5 3.25 2.05

(sd) 3.4 1.4 1.65 1.32

One-sided celebrity (n) 19 20 20 20

(x) 8.18 3.55 3.75 2.7

(sd) 5.09 1.57 1.55 1.26

Two-sided non-celebrity (n) 19 20 20 20

(x) 8.5 5 3.53 3.18

(sd) 5.72 1.27 1.74 1.63

Two-sided celebrity (n) 17 17 17 17

TABLE 2

ANOVA results for effects of sidedness and spokesperson type

a indicates significance at p < 0.05.

References

Z. Erdogan and P. Kitchen, ‘Getting the best out of celebrity endorsers’, Admap (April 1998).

M. Kamins, M. Brand, S. Hoeke and J. Moe, ‘Two-sided versus one-sided celebrity endorsements: the impact on advertising effectiveness and credibility’, Journal of Advertising 18(2) (1989): 4–10.

Michael A. Kamins, ‘Celebrity and noncelebrity advertising in a two-sided context’, Journal of Advertising Research (June–July 1989): 34–42.

Michael A. Kamins, ‘An investigation into the “match-up” hypothesis in celebrity advertising: when beauty may be only skin deep’, Journal of Advertising 19(1) (1990): 4–13.

George Lazarus, ‘Tiger’s shooting for an endorsement record, too’, Chicago Tribune (5 May 1997): 4, 6.

Al Ries, ‘Count on consumers to follow the leader’, Brandweek 38(25) (23 June 1997): 18.

Questions

What kind of marketing research could businesses conduct to determine whether their products would perform better with celebrity endorsements?
Discuss the role of multi-dimensional scaling in the matching of a celebrity to the right product.
Could conjoint analysis be used to determine whether celebrities should be used and, if so, which celebrity should be selected? How could it be used?
What kinds of precautions or pre-testing should the research engage in to ensure that the celebrities and two-sided advertisements used in the experiment were appropriate? What complications or contaminations might be present in the experimental results if these precautions were not taken?
Based on the results presented, do two-sided advertisements have an advantage over one-sided advertisements? Celebrity advertisements over non-celebrity advertisements?
Which type of advertisement is the most effective? The least effective? (Hint: look at the ANOVA results.)
Is analysis of variance an appropriate technique to use to analyse the data obtained in this study? Why or why not?
Could regression analysis be used to analyse the data obtained in this research? If so, how?

CASE 4

The demographic discovery of the decade

Many marketers are ill positioned to take advantage of the most important consumer segment groups to rise over the next 20 years, and beyond – senior citizens. It is important that marketers and advertisers do not fail to connect with their older audiences. Not only are baby boomers holding a large proportion of economic wealth – 80% of all financial wealth in the UK and Canada, and over 50% of discretionary income in the USA – they are also major buyers of luxury products such as cars, alcohol, vacations and financial products. Nonetheless, marketers continue to aim promotions at and cater to younger segments.

Critics cite an eagerness to use mainly young characters in advertisements, and a tendency to portray old age as undesirable, as evidence of advertisers’ ageism. In general, age discrimination has been given a low priority, but this is changing as demographics demonstrate growing numbers of older people in the population who have reason to protect and promote their value in society. One reason suggested for the disinterest shown towards older people is the youthful profile of many of those working in advertising. The average age of most advertising executives is below 50, and many researchers have commented on the lack of empathy this seems to create with the over-fifties population. Research has found the average age of US advertisers’ representatives to be 31, and the average age of agency representatives to be 28. A similar picture emerges in continental Europe, where executives are also in their twenties and thirties, implying a disconnection between agency demographics and those of the marketplace.

The reluctance of the marketing industry to target the older population is in part due to outdated and stereotypical notions of age and ageing. The over-fifties today form a very different profile from the ‘elderly’ of the past; healthier diets, improved life expectancy and a widespread desire to feel younger for longer mean that they neither look nor feel old, and certainly do not want to be viewed as such by marketers and advertisers. A lack of empathy among many marketers and advertisers for the needs of older consumers has manifested itself in communications that are often inappropriate for this new generation of over-fifties.

Senior citizens should not be seen as one homogeneous mass. As with all types of consumer, there are many ways to segment senior citizens, one being based upon age. Using ‘age’ as a segment descriptor, senior citizens can be divided into four segments. There are the so-called older adults ranging from 55 to 65 years of age. The second market segment, the elderly, is made up of those aged 65 to 74. The aged, those from 75 to 84, and the very old, those 85 and over, constitute the other two segments.

A closer look at the older adult group reveals that they are interested in maintaining a youthful appearance and are major targets for exercise equipment, health programmes, diets, cosmetics, cosmetic surgery, sports clothing, designer wear, and a wide array of personal services that improve appearance. An increasing number of older adults opt for early retirement or move into new careers and part-time jobs. The elderly group comprises those who have been retired for some time. They tend to take a keen interest in health and nutrition and to be concerned with diet, salt intake, cholesterol, fried foods, and calories. They often drink less alcohol than the younger population and are a good market for skin care products, prescriptions, vitamins and minerals, health and beauty aids, and medicines that ease pain and promote the performance of everyday activities. The aged group often has health and mobility problems and hence requires health care services and special care facilities. The very old need help in their day-to-day tasks. They find it difficult to get around and need regular medical and hospital care. Again, they represent a large market for health care facilities.

While the classification of the mature market into these four segments has been useful, another classification, and perhaps a better one for advertising purposes, is based on attitudes towards advertising. These segments could then be profiled in terms of psychographic variables. A major concern of advertisers targeting the aged consumer has been the way in which the older population utilises and evaluates information from advertising to make purchasing decisions. One study by Davis and French explored aged consumers’ use of advertising as a primary source of information in purchase decisions. The respondents were clustered based on attitudes towards advertising. Psychographic profiles were developed for each of the derived segments.

A database of annual lifestyle surveys was used to obtain a sample of 217 married female respondents aged 60 and over who were not employed outside the home. Respondents were asked to rate their degree of agreement with each of the 200 AIO (activities, interests, opinions) statements on the survey. Respondents were also asked to rate four attitudinal statements measuring information usage and beliefs about advertising, as well as the credibility of the source of advertising. Identical information obtained from a previous study was used for replication purposes by Davis and French.

The data on the four statements (shown in Table 1) measuring attitudes towards advertising were analysed used Ward’s method of clustering. Three clusters — Engaged, Autonomous, and Receptive consumers — were identified. Mean scores for each cluster are presented in Table 1. To test stability, replication of the cluster analysis was undertaken using the data obtained in the previous study. Ward’s method of clustering was used to analyse the data from the previous study. Again, three clusters were obtained. Cluster means obtained by Davis and French on each of the clustering variables for the replication sample (previous study) are also shown in Table 1.

To determine the psychographic differences among the three clusters, two additional steps were taken. First, one-way ANOVA was carried out to determine the discriminating variables. The three segments formed the grouping or the independent variable, and each psychographic statement served as a dependent variable. Forty-one of the original 200 psychographic statements were found to be statistically significant. With the realisation that some of these significant variables were probably measuring the same characteristics, a principal components factor analysis was carried out, with four factors (accounting for 60.3% of the variance) extracted in a varimax rotation. Factor scores were computed for each of the three segments by Davis and French, and Table 2 shows these scores, along with the variables that loaded highly on these factors and the variable means. This information can be used to provide psychographic profiles for each of the three segments identified in cluster analysis.

Table 1

Table 2

References

M. Carrigan, and I. Szmigin, ‘The representation of older people in advertisements’, Journal of the Market Research Society 41(3) (July 1999).

B. Davis and W. French, ‘Exploring advertising usage segments among the aged’, Journal of Advertising Research (February–March 1989): 26.

‘Seniors in stores’, American Demographics 18(4) (April 1996): 44.

‘The ungraying of America: although population is getting older, today’s older consumers are not necessarily acting same way older consumers behaved in the past’, American Demographics (July 1997): 12.

See also

Questions

Studies have found that the older, elderly, aged and very old segments of the mature market need good supporting health services and facilities. Describe in detail how health maintenance organisations (HMOs) can effectively determine the differences in the health care needs of these segments. What kind of information should be obtained? Which statistical techniques should be used to analyse the data?
Do you think that the data analysis strategy adopted in the study reported in this case was appropriate? Why or why not?
Qualitatively describe each of the three clusters, based on the information in Table 1.
Interpret each factor in Table 2.
Do you think that the study reported in the case should have used discriminant analysis? If so, how?
Suggest an alternative data analysis strategy for the study reported.

CASE 5

DuPont has designs on fashion

DuPont () Fibres Division was wondering: could carpets move ‘up-market’ into the fashion-conscious world typically associated with clothing and furniture? The long-time market share leader in the carpet industry, DuPont was searching for new ways to expand in a slow growth market, which had been growing at approximately 5% per annum over the period 1990–7 and was predicted to remain level or even decline in following years. Earnings were down by 1.7% as of 31 March 1998 for DuPont fibre operations, and the carpet industry was flat. In addition, in the residential segment of the market, DuPont needed a way to differentiate its nylon fibre carpets from increased competition, including the strong challenge of Amoco’s new polypropylene fibre carpets.

The carpet industry can be divided into three end-use segments:

Commercial carpets for offices, hospitals, hotels, schools, government facilities and industrial sites
Contract residential carpets for large residential purposes like apartment complexes or subdivisions
Residential carpets for homes.

DuPont estimated that 30% of carpet sales were from new residential construction, and the other 70% were from replacement purchases. In addition to different end-use segments, each segment’s consumers were quite different from the other segment’s consumers and each segment was serviced through different channels. Interior designers, architects or specifiers for their clients typically bought commercial carpets. They had to be durable and, especially for some end-uses like hotels, fashionable.

Contract residential carpets were bought by designers or by the contractor who typically focused on price and durability. The residential segment differed from the other two because the purchaser was also the end-user. Typically, the householder purchased carpeting from a retail outlet in order to create the atmosphere wanted in their home. In a recent trade journal article of Buildings, carpet industry experts were quoted as saying that ‘retail stores (are) an important and effective component of the retail (environment)’. Ninety-seven percent of all carpeting is produced from man-made fibres derived from petroleum. The leading fibre in the industry has been nylon, which was invented by DuPont in the 1930s. Though the leader in the industry, DuPont was challenged by other high quality nylon producers, like Monsanto, Allied, and BASF, and a host of generic low-end fibre producers. DuPont had 27% of the global nylon market and 58% of the nylon-6/6 market (the most advanced nylon fibre).

Fibre producers sell their output to carpet mills that then produce the carpeting. Until the 1980s, competition in the residential segment was based on the technical qualities of the fibres, the mill price of fibre, and reliability in shipping, none of which directly affected the consumer. As such, the industry tended to be very production oriented. In the early 1980s, a major breakthrough benefiting the consumer emerged, stain resistant carpets. By applying a chemical coating to the carpet fibre during production, the carpet was protected from permanent staining arising from most household soiling agents. Stains could be wiped off the carpet, thus alleviating the concern many people had about entertaining or ‘living’ on their carpet. The four major fibre producers quickly announced their versions of the stain resistant carpets in an effort to remain competitive.

Amoco also entered the market when it announced a ‘new revolution in carpeting’, carpets made from polypropylene (PP) fibres rather than nylon. As a fibre, PP is inherently stain resistant, thus it offered the best overall protection from stains and it costs less to produce than nylon. However, it did not as readily accept dyes as nylon, nor was it as soft to the touch, thus making it less fashionable. As such, PP initially had difficulty entering the residential segment, but was well received by the commercial segment. Approximately 24% of the total US carpet fibre market is PP.

In the mid-1980s, technical advancements in dyeing allowed Amoco to seriously compete in the residential segment. Their objective was to lead the introduction of PP fibres to the residential market and they were strongly committed to doing just that. DuPont, who desired to be the unquestionable leading fibre producer to the residential segment and relied more on marketing than other companies, seemed to sense the threat of PP towards nylon fibres when it purchased Hercules fibres, a large PP extruder. However, DuPont believed that nylon was still the fibre for residential carpets and was determined to let the consumer cast the deciding vote.

DuPont believed that the consumer needed to be given a greater role in the carpet industry and it believed it could use its well-respected company name to attract consumers. As such, DuPont created the first fibre-producer backed carpets, DuPont Stainmaster carpets, which carried a guarantee backed by DuPont on stain resistance, wear and anti-static, and was branded as a DuPont carpet. The results of the programme were highly successful with DuPont creating high brand recognition among consumers, the first time this had ever been achieved by a fibre company, thus differentiating it from the other fibre producers. DuPont took its Stainmaster carpet business and converted it to a premium carpet brand by increasing prices and reducing its distribution channels. DuPont spent over $10 million on television advertising beginning in 1995. A survey conducted by Video Storyboard Tests, Inc. rated the DuPont Stainmaster television commercial the fourth ‘most popular television commercial’.

Based on its success with DuPont Stainmaster, the industry leader decided to forge new ground. For most of its history, the residential segment had typically been the most blasé segment of the carpet market. Styles tended to be simple, colours passive, and features uniform across all competitors in the industry. Technically, industry players maintained that differences did indeed exist, but in the words of one industry analyst, ‘The differences were there in style and fibre quality, but the householder out shopping for carpet did not really know or care, they only like what they could see and feel.’ As such, DuPont wondered if the styles and designs so popular in the commercial segment could be transferred to the residential segment. Was the average household willing to make carpeting more than just a backdrop for other furnishings? If the programme were to be successful, it would mean further differentiation from other nylon fibre producers by creating a new segment based on fashion and status for residential carpets, increasing brand awareness among consumers for DuPont (resulting in increased demand for DuPont fibres), and bringing in a new dimension to residential sales which would be difficult for PP carpets to duplicate and would solidify DuPont as the leader in the residential segment. However, it would be an expensive undertaking, involving a significant amount of publicity and risking its reputation with its major fibre clients. In addition, if DuPont were to proceed with the Designer Collection, it would have to move fast. The largest trade show in the industry, in which the companies announced their major designs and programmes for the coming year, was due in three months. If DuPont wanted the Designer Collection to have a maximal impact, it would have to have carpet samples and promotional materials ready for distribution at the show, and their marketing programme in line to begin shortly thereafter. Thus, the go or no-go decision would have to be made within six weeks.

References

DuPont 1997 Annual Report.

DuPont Quarterly Report, 31 March 1998.

Gregory Morris, ‘DuPont Canada brings on nylon-6/6 expansion’, Chemical Week (29 April 1998): 19.

Marc Reisch, ‘New texture in carpet fibres’, Chemical & Engineering News 76(4) (26 January 1998): 20–1.

Andrew Wood, ‘DuPont wants to hitch up nylon’s performance’, Chemical Week (29 October 1997): 42.

www.dupont.com

Questions

Chapter 1: Introduction to marketing research

Marketing research involves the identification, collection, analysis and dissemination of information. Explain how each of these phases of marketing research applies to DuPont's problem.
Is the problem facing DuPont a case of problem identification research or problem solution research? Explain.

Chapter 2: Defining the marketing research problem and developing a research approach

What is the management decision problem facing DuPont?
What is the marketing research problem facing DuPont?
Break down the general marketing research problem statement into component parts.
What theoretical findings can assist in developing an approach to the problem?

Chapter 3: Research design

Can exploratory research be used in this case? How?
Can descriptive research be used in this case? How?
Can causal research be used in this case? How?

Chapter 4: Secondary data collection and analysis

What published sources of secondary data can you identify which would be helpful?
How may the Internet be of help in tracking down secondary data and intelligence?

Chapter 5: Internal secondary data and the use of databases

What internal sources of secondary data can you identify which would be helpful?
Which computerised databases can be used? What is their biggest disadvantage?

Chapter 6: Qualitative research: its nature and approaches

What makes the word ‘designer’ a qualitative issue?
How may an ethnographic approach help to understand how consumers choose carpets?

Chapter 7: Qualitative research: focus group discussions

Develop a moderator's outline for a focus group to assess consumer desires in residential carpets with respect to the Designer Collection.
DuPont has hired a marketing research firm to conduct focus group sessions. What are the ethical considerations of DuPont managers sitting in on these focus group meetings?

Chapter 8: Qualitative research: depth interviewing and projective techniques

What potential would there be to use depth interviewing in this project?
Design sentence completion techniques to uncover underlying motives.

Chapter 9: Qualitative research: data analysis

How may the social and cultural background of qualitative researchers, moderators or interviewers affect how they pose questions and interpret answers?

Chapter 10: Survey and quantitative observation techniques

Match the criteria for selecting survey methods with the survey method(s) offering the best results for DuPont.
Which survey method would you recommend to DuPont to conduct descriptive research? Why? What are the limitations of this mode?

Chapter 11: Causal research design: experimentation

Based on the DuPont project, give an example of each of the conditions of causality for the relationship between purchase of designer carpets and income level.
Is causal research necessary in this case? If so, which experimental designs would you recommend and why? If not, devise a scenario in which causal research would be necessary.
What extraneous variables are threats to the internal and external validity in the design you have selected?
Can a field experiment be used to conduct the test? Explain.

Chapter 12: Measurement and scaling: fundamentals, comparative and non-comparative scaling

What types of non-comparative scales can be used to gather the information needed on psychographics, motivations, attitudes and intentions?
In designing scales for the survey, which scales do you recommend?
How would you determine the reliability of the scales?
How would you assess the validity of the scales?

Chapter 13: Questionnaire design

Are each of the following questions well formulated? If not, what is the error?

a. What is your favourite construction of carpet fibres?

Nylon BCF __________

Nylon Staple __________

Polypropylene BCF __________

Polypropylene Staple__________

Polyester __________

b. What style of carpeting do you have in your office?

Uniform colour; Conservative Style __________

Uniform colour; Fashionable Style __________

Multicolour; Conservative Style __________

Multicolour; Fashionable Style __________

c. Do you intend to buy a new carpet soon?

Yes __________

No __________

d. Do you believe, as most Europeans do, that European citizens should buy European-made carpets?

Yes __________

No __________

e. Will you buy designer carpets given that they cost slightly more than traditional carpets?

Yes __________

No __________

Design a questionnaire to be used in a survey.

Chapter 14: Sampling: design and procedures

Answer questions 33 through to 36 assuming that an in-home interview is being conducted.

What is the target population for this study?
What sampling frame can you use?
What sampling technique do you recommend for this study? Why?
What non-response issues must be considered and how can they be overcome?

Chapter 15: Sampling: final and initial sample size determination

Suppose DuPont conducts a preliminary market study of 30 respondents to determine the price they are willing to pay for carpets in the Designer Collection. The mean response is calculated to be €30.00. If DuPont wants to be 99% sure that the true value lies within €1 of this figure, how large a sample do they need to survey given that the population standard deviation is €5.00?
Suppose DuPont wants to know how many households are interested in the Designer Collection. To do so, they conduct a pilot study and learn that 21 of 30 respondents expressed an interest in designer carpets.

a) How large a sample does DuPont need to draw in order to be 99% sure that this result is within 5% of the true value?

b) What if they wanted it to be only within 20% of the true value at a 99% level of confidence?

c) What if DuPont only required an 80% level of confidence at a 20% level of precision?

Chapter 16: Survey fieldwork

Answer the following questions assuming an in-home interview is being conducted.

What characteristics would you look for when hiring fieldworkers for this survey?
What issues are most important in training your fieldworkers for this survey?
What issues must you as the supervisor be most concerned with during the interviewing?
How would you validate the fieldwork?
How would you evaluate the success of your fieldworkers?

Chapter 17: Data preparation

Suppose, after receiving completed questionnaires, the following results were obtained from three different respondents: respondent one consistently used the lower end of the scale, respondent two consistently used the middle of the scale and respondent three consistently used the upper end of the scale. Correct for the response style of the respondents in order to ensure meaningful results on Q6. Assume the following means and standard deviations.

Chapter 18: Frequency distribution, cross-tabulation and hypothesis testing

Suppose you administered the survey designed in Chapter 13, and you collected data from 240 respondents (see Appendix for details). Run the following analyses on the data and draw conclusions from the results obtained.

Run descriptive statistics and obtain frequency distributions for all variables.
Cross-tab Q3 (Is carpeting important?), Q4 (Is carpeting fashionable?), Q5 (Is carpeting central?) and Q6 (Is carpeting durable?) with the demographic variables in Q20 to Q23. If results are poor, you may have to create dummy variables in order to obtain valid results.
Conduct a t-test for each of the seven attributes listed in Q7 (Importance ratings of attributes) by each of the four responses to Q2 (What carpet do you own?), i.e. there will be 28 t-tests run.

Chapter 20: Correlation and regression

Question 45 continued:

Regress each of the four styles of designer carpets in Q9 (Rating for styles) on the seven attributes of Q7.
Regress each of the styles in Q9 on all attributes in Q10 (Desirability of attributes), i.e. there will be four regressions.
Sum the seven attributes in Q10 to get one score for each respondent. Then use the stepwise procedure to regress this variable on Q15 (AIO items).

Chapter 21: Discriminant analysis

Question 45 continued:

Conduct four two-group discriminant analyses on Q9 (a–d) using the attributes in Q10 as the independent variables. For each style A–D in Q9, group the respondents as low or high on their rating of the style. Let a low rating be 1–4 and a high rating be 5–7. Use this new variable as the dependent variable in your discriminant analysis.

Chapters 22–23: Factor analysis, Cluster analysis

Question 45 continued:

Run a factor analysis of Q15 and cross-tabulate the factor scores with the responses in Q2. Cluster the results of the factor analysis into three clusters in order to segment the respondents on psychographic measures. Run cross-tabulations on the cluster centres to determine whether any relationship exists between the clusters and the current ownership of carpets, Q2.

Chapter 26: International marketing research

Due to the sluggish nature of the domestic carpet market, DuPont wants to explore the possibilities of moving into international markets. Before they design an international marketing research study, however, they want to determine which end-use segment (commercial carpets, contract residential carpets or residential carpets) would offer the most potential, and just concentrate their marketing research efforts on this segment for their initial foray into the international arena. How should they do this?

Acreman, S. and Pegram, B. Research, November 1999, pp. 36–41.