Each abstracted article is awarded 0–5 stars for each of four qualities:

1. depth of research
2. value in practice
3. originality of thinking
4. Readability for non-specialists.

No abstract is included for any article awarded less than seven stars overall.

Affiliate marketing

M. Downs, J. York, B. Clayton, C. Tradgett, D. Hall and L. Gregoriadis

EXPLANATORY. MultiChannel Marketing (UK), February–March 2007, pull-out supplement (12pp)

Affiliate marketing, affiliate networks, commission

This collection of short articles discusses the growth and potential of affiliate marketing (now accounting for 10 per cent of online sales) as a means of driving traffic to a merchant's website. Covers use of this channel in B2B (mostly for SMEs) and in retail. Discusses DIY systems (eg Amazon) and advantages of using networks (either one or more), which supply tracking software, handle payments, etc. Describes affiliates as a form of 'commission only' salesmen. Reveals that commission payable to affiliates may vary from 1 to 25 per cent, but in retail is typically 4–5 per cent, with the network, if used, taking a further 30 per cent of that. Emphasises the need to use complementary affiliates, rather than using a scatter-gun approach. Notes the need to keep affiliates supplied with up-to-date and inspiring data about your company and products. Lists key questions to ask a network provider. Discusses use by affiliates of brand bidding on merchants' names. Tabulates 16 affiliate networks, with contact details and brief descriptions.

These articles are not well coordinated; however, if you persevere, they provide a valuable introduction to a burgeoning new channel.

Research:** Practice:*** Originality:* Readability:***

Ref: 9301

Fulfilment

M. Bottomley, L. Thompson and D. Erskine

EXPLANATORY. MultiChannel Marketing (UK), August–September 2007, pull-out supplement (10pp)

Fulfilment, outsourcing, offshoring, SOA, QA

This collection of short articles deals with the 'unglamorous' business of fulfilment services, which have become increasingly significant with the growth in e-tailing, and in particular with the issues surrounding outsourcing of such services, from contact centres to order processing and warehousing. Claims that the move to relocate call centres abroad ('offshoring') has recently started to reverse, due to adverse customer reaction impacting some high-profile companies. Suggests this will continue. Lists a number of reasons why companies may prefer to handle fulfilment activities in-house. Suggests a major motive for outsourcing remains the need to cope with fluctuating — perhaps seasonal — volumes. Discusses the advantages of shared systems between client and provider. Recognises this may not always be feasible; discusses distributed systems using applications from different suppliers through Service Oriented Architecture (SOA). Gives a checklist for those trying to select an outsourcing partner; discusses quality assurance; lists four suppliers in the field.

Nothing new for veteran direct marketers, but a useful introduction for aspiring e-tailers to a subject too often overlooked as boring.

Research:* Practice:*** Originality:* Readability:***

Ref: 9302

The 'edgy' 2012 logo has abolished Olympian grace and geography

I. Jack

JOURNALISTIC. The Guardian (UK), 9th June, 2007, p. 32 (1pp)

Branding, logos, Wolff Olins, Olympics, 2012, Interbrand, BA, London

Compares the Wolff Olins logo for the 2012 Olympics with the image used in London's last Olympiad in 1948. Notes widespread public distaste, and mockery, for the new logo, which it claims has no geographical reference, and no grace; it is intended to appeal to everyone, regardless of age, culture and language. Compares this ambition with that surrounding BA's catastrophic attempt to change its logo in much the same way. Notes that brand professionals tended to see the BA case as a customer 'losing his bottle'; quotes the chairman of Interbrand on the need to show 'real united confidence — absolutely no hesitation or wobble; people get used to things'. Concludes that global brands 'are the enemy of the local and erode our...sense of belonging'. Contrasts the human attachment to things past with the impatience of branding experts to create the future.

The chairman of Interbrand sounds like the sort of ship's captain that would rather go down with her ship than admit that it had hit an iceberg. Whatever your views on the 2012 logo, this short article raises some interesting questions about the place and purpose of branding and logos today.

Research:* Practice:** Originality:** Readability:****

Ref: 9303

The role of internal branding in the delivery of employee brand promise

K. Punjaisri and A. Wilson

SURVEY. The Journal of Brand Management (UK), Vol. 15, No. 1, p. 57 (15pp)

Corporate branding, corporate marketing, employees, Thailand

Notes the rise of corporate branding, and corporate marketing, and the recognition of the crucial role played in these by employees. Shows a diagram of Balmer's corporate marketing mix, one leg of which is corporate brand management, with which this article deals: multiple stakeholders interacting with employees. Notes that research into processes required to encourage brand-supporting behaviour is limited, as is research into the attitudes of employees themselves. Describes a study of attitudes at six 4- and 5-star Thai hotels, comprising 1½ hour interviews with 20 managers and 30 customer-facing staff to produce qualitative data, followed by a questionnaire completed by 699 staff to produce quantitative data. Considers the extent to which internal communications (daily briefings, newsletters, logbooks, notice boards) and training were considered major mechanisms in internal branding: finds that ICs have a stronger effect than training. Considers HR's problems in recruiting the 'right' people in the first place, and the use of a probationary period.

It makes a change to look at problems of internal branding through the eyes of employees: this is a good departure in principle, though a little light in findings.

Research:*** Practice:** Originality:** Readability:**

Ref: 9307

'i-Branding': Developing the internet as a branding tool

G.J. Simmons

THEORETICAL. Marketing Intelligence & Planning (UK), Vol. 25, No. 6, p. 544 (9pp)

Branding, i-Branding, Four Pillars

Briefly introduces the concept of branding; contends that the internet, by increasing the power of the consumer relative to the producer, has rendered branding a more complex and dynamic exercise; little integration of i-branding is yet evident. Introduces the concept of the Four Pillars of Branding: Understanding customers; Marketing communications; Interactivity; Content. Claims that a successful brand has three aspects: it is dependent on customer perception, which is influenced by added-value characteristics, which must be sustainable. Goes on to elaborate on each of the Four Pillars: Understanding customers — lists ways of acquiring customer data, such as server-side data capture, client-side data capture, online surveys, databases; Marketing communications — using an internet presence on a website to encourage relationships and create mutual value; Interactivity — through email, viral marketing, blogs, RSS feeds, online communities; Content — deals with such issues of site design as multimedia, graphical user interfaces, entertainment values, etc. Emphasises the need to integrate the Four Pillars: marketing communications have to be conducted with an understanding of customers (permission marketing; no spamming); content must be based on what the customer requires, not on how the company is structured; interactive content will encourage return visits.

A contribution to the ongoing debate on the place of branding in the internet era.

Research: - - - - - Practice:** Originality:** Readability:***

Ref: 9315

If brands are built over years, why are they managed over quarters?

L.M. Lodish and C.F. Mela

THEORETICAL. Harvard Business Review (US), July 2007, p. 104 (11pp)

Brands, discounting, advertising

Notes the growth, between 2003 and 2005, of the private label market by 13 per cent, and a 50 per cent increase, over the last 25 years in consumer price sensitivity. Contends that brands are on the wane; claims that much of this change is due to the increasing adoption of a short-term view in favour of discounting and promotions in preference to the long-term investment in advertising, new product development and new forms of distribution. Blames this situation on the increasing availability of sales data — instancing scanner data in particular. Claims that from 1978 to 2001 trade promotion spending grew from 33 to 61 per cent of firms' marketing budgets, while advertising spend fell from 40 to 24 per cent. Notes, as contributory factors, that long-term effects are harder to measure, and that brand managers typically have short tenures — often less than a year. Advocates greater attention to long-term dashboard data, including data on quantity and price premiums. Quotes, among other examples, the case of the Clorox bleach product, which rescued a dire situation by cutting discounting and increasing advertising, enabling it also to increase prices, and revenue and profit.

An interesting analysis—but in looking at reasons for a decline in brand power it neglects to consider the effects of the internet, and the shift in producer/consumer power relationships that accompanies e-retailing.

Research:*** Practice:*** Originality:** Readability:****

Ref: 9317

Roots marketing: The marketing research opportunity

C. Nancarrow, J. Tinson and R. Webber

THEORETICAL. International Journal of Market Research (UK), Vol. 49, No. 1, p. 47 (23pp)

Ethnicity, ancestry, migrants, Scottish, ex-pats

Sets out to explore the extent to which there may be a link between personal consumption habits and ethnic/national origin or ancestry, particularly among persons not resident in their country of birth/ancestry. Takes, as a basis for this study, the case of persons of claimed Scottish birth/ancestry now living in England. Notes that Scots nationality, unlike some other origins, cannot in general be inferred (except by accent, which may fade and is unlikely to be present in generations following migration). Notes the difference between those non-English persons who elect to display their distinctiveness, and those whose distinctiveness is self-evident and, in that sense, forced upon them. Describes a series of three studies of persons claiming Scottish birth or descent (through 1–3 generations). Notes the appearance of two characteristics among respondents: an identification with Scotland through a 'sense of place' and/or through a 'sense of tribe'. Quantifies the extent to which a sense of Scottishness is greater among ex-pats, and how purchase of Scottish products gradually declines over the generations, while still greater than for persons not of Scottish descent. Concludes with a discussion of how migrants can be reached by marketers.

Opens up a whole series of intriguing possibilities for niche marketing—or perhaps just for more intelligent segmentation of existing markets.

Research:**** Practice:** Originality:***** Readability:***

Ref: 9304

Helping bank customers switch: A case study

C. Matthews and D. Murray

CASE STUDY. Journal of Financial Services Marketing (UK), Vol. 11, No. 4, p. 360 (10pp)

Banks, switching costs, New Zealand

Notes that among the six main banks in New Zealand, plus some non-bank financial institutions, there is a low customer churn rate of some 4 per cent p.a. In a mature market, this means a problem for those wishing to grow. Speculates this may be due to high switching costs. Categorises switching costs as Procedural (time and effort required), Financial and Relational, and lists eight costs under these heads. Describes a pilot project by one institution designed to ease/eliminate these costs for new customers, which ran for two months, in two branches, in 2001. This pilot was not advertised. The institution waived establishment fees for new customers, and undertook to deal with customers' previous banks in closing/transferring their accounts. Reports the results of this pilot: some new customers did not have a previous bank, others did not want to close existing accounts with other banks; others preferred to do the job themselves. The remaining few caused unanticipated amounts of trouble; time spent on opening each new account rose from under an hour to half a day. Further, the banks where accounts were to be closed made things worse through lack of cooperation and/or incompetence; the pilot was abandoned.

This seems to have been a naïve exercise; banks, which are frequently unable/unwilling to provide proper service to loyal customers, are certainly not going to exercise themselves on behalf of the disloyal. Still, a chilling example of the negative power of switching costs.

Research:** Practice:* Originality:** Readability:***

Ref: 9305

The distance marketing of financial services — A UK overview

W. Yonge and R. Massey

LEGAL SURVEY. Journal of Financial Services Marketing (UK), Vol. 11, No. 4, p. 370 (11pp)

EU Directives, legislation, financial services, distance marketing

Notes the exponential growth of distance marketing and contracting in financial services, and the consequent growth in regulatory instruments. Lists the three European Directives on which UK legislation is based: Distance Marketing of Consumer Financial Services Directive (DMD); Data Protection Directive and E-Commerce Directive. Describes how implementation of the DMD (which applies only to B2C marketing) is divided between the FSA, the DTI and the Treasury. Describes the information requirements (with exceptions for telephone contracts); customers' rights of withdrawal; position of consumer credit agreements; ban on inertia selling. Deals with the provisions of the Privacy & Electronic Communications Regulations (which implement the E-Commerce Directive) regarding unsolicited communications, country of origin considerations, cookies and spam. Deals briefly with the Data Protection Act, the export of personal data beyond the EU and the Safe Harbor Principles.

A very thorough run-down on what reads like a legal minefield, but is not, in the end, much more than good practice.

Research:** Practice:*** Originality: - - - - - Readability:***

Ref: 9306

How similar are frontline bank employees' perceptions of service quality to their customers'? A study of female customers and employees in Turkey

U. Yavas

SURVEY. Journal of Financial Services Marketing (UK), Vol. 12, No. 1, p. 30 (10pp)

Banking, customer service, QOS, Turkey

Notes the growing realisation in the banking world of the requirement for quality of service to customers, and the importance of front-line staff in achieving this. Describes surveys conducted with two groups (151 students and 156 adults) of female bank customers, and 68 female front-line bank employees. Lists 22 factors by which customers' perceptions of quality of service were measured on a seven-point scale (from much worse to much better than expected). Indicates that employees were asked to put themselves in customers' positions before answering questions (again on a seven-point scale from strongly agree to strongly-disagree) related to the above service quality items. Reveals that there was good congruence of views between the two groups of customers, but poor congruence between these and the views of bank employees. Concludes that bank staffs must learn to look at, and define, service quality as perceived by the customer, and not simply as defined by the bank or its employees.

Read the first sentence of this abstract carefully again: isn't it rather astonishing? Still, better late than never. Is Turkey perhaps unusually slow in this area? I beg leave to doubt it.

Research:*** Practice:** Originality:** Readability:**

Ref: 9308

A marketing paradox

M.E. Hill and J. McGinnis

VIEWPOINT. Marketing Intelligence & Planning (UK), Vol. 25, No. 7, p. 652 (9pp)

Marketing, marketing thinking, academic/practitioner divide

Describes the growth of the marketing discipline, and of marketing knowledge since 1900. Suggests that there is a paradox (described as the 'contributing-hindering', or C-H paradox), which arises from concentrating on the development of accepted theories and principles, which become ever more entrenched, thereby stifling new thinking. Suggests that this paradox cannot be resolved, but only understood so as to work round it. Notes that concepts developed almost half a century ago still hold sway in marketing circles, although the context in which they operate has changed radically (AIDA, the marketing mix, the 4Ps). Notes the growing academic/practitioner divide: the complete absence of practitioners from the boards or reviewing panels of three leading marketing journals in the USA, UK and New Zealand. Wishes to redefine marketing as being about thinking, rather than just about explaining and doing, involving creative thinking, critical thinking and reflective thinking. Suggests that theoretical thinking might well begin with practitioners: what questioning does their perspective require? Or by discovering from consumers what questions they ask in navigating the marketplace. Notes a number of obstacles to marketing thinking, culminating in scarcity of time — particularly for practitioners. Claims that theories and concepts are the product of thought, not themselves thinking, but starting points for thinking, and that a practitioner solving a practical problem contributes as much to marketing thinking as an academic working on a strategic theoretical problem.

From a practitioner's point of view this article is making a lot of the right noises, and is worth reading. But one might wish for a few practical examples of the type of new thinking called for. Perhaps that's up to the readers.

Research:** Practice:* Originality:*** Readability:***

Ref: 9309

Information overload: A cross-national investigation of influence factors and effects

C. Klausegger, R.R. Sikovics and Huan Zou

SURVEY. Marketing Intelligence & Planning (UK), Vol. 25, No. 7, p. 691 (18pp)

Information, information overload

Sets out to examine the nature and negative effects of information overload in the context of the rapid development of new information technologies. Notes that information is now regarded as a production factor alongside the classic trinity of land, labour and capital. Seeks to investigate the factors influencing information overload, and its effects on individuals. Lists previous studies of the phenomenon. Notes that information overload has both an objective and a subjective element. Notes that the collection, transformation, storage and transfer of information is estimated at more than half of overall costs in US businesses (rising to near 100 per cent in information-driven companies): information is replacing tangible assets such as inventories, and the ROI on investment in IT has been found to be higher than for other capital goods. Notes the growing pace of change as a factor in increasing the need for information. Lists seven reasons why managers collect so much information, and six reasons why a piece of information is highly valued. Describes an empirical study conducted by telephone with 511 UK company executives, 502 US and 100 each in Australia, Hong Kong and Singapore, who were asked to assess the incidence and effects of information overload on their colleagues, now and in the expected two years ahead. Shows tables indicating the sources on information overload, the differences between these countries and the methods of adapting to the problem used by individuals, including delegation, postponement, working overtime and taking work home.

This study was done in 1996, although written up (after confidentiality problems) only in 2007. Certainly the identification of the internet as a cause of information overload would have been much greater ten years later, but perhaps the real lesson is that perceived information overload has always been with us. But it still does need to be taken seriously by management.

Research:*** Practice: ** Originality:** Readability:***

Ref: 9310

Role of electronic trust in online retailing: A re-examination of the commitment-trust theory

A. Mukherjee and P. Nath

THEORETICAL. European Journal of Marketing (UK), Vol. 41, No. 9/10, p. 1173 (30pp)

Online retailing, trust, commitment

Notes the growth of online retailing, but the widespread reporting of cases of insecurity, fraud, spamming, etc, all leading to a lack of trust, and inhibitions for online purchases. Looks at commitment–trust theory as developed by Morgan and Hunt, and asks whether it needs modification in an online environment. Looks at the antecedents of trust and commitment, and their effects on customer behaviour. Notes the ways in which online retailing differs even from other forms of distance selling — notably the different level of interaction. Identifies as key components of trust: customer propensity to trust the retailer, confidence in the website, trust in internet technology. Lists five antecedents of trust: shared (ethical) values (eg permission marketing); communication (ie openness, quality of information and of response); fear of opportunistic behaviour; privacy; security. Lists three antecedents to commitment: relationship termination cost; relationship benefit and shared values. Tabulates all of the above variables, together with appropriate measures for their optimum resolution. Gives a series of 11 hypotheses about the effects of these antecedents on consumer behaviour; describes an online survey to test these hypotheses Finds a strong positive relationship between trust and commitment; privacy, followed by security are the most important elements of trust. Termination cost is negatively related to commitment.

The findings are far from startling, but to have them presented in a structured manner which allows them to be addressed one by one—as they should be—by online retailers, is beneficial

Research:*** Practice:** Originality:** Readability:***

Ref: 9311

Mistaking demographic segments for people: Another source of customer abuse

H.J. Rotfeld

VIEWPOINT. Journal of Consumer Marketing (UK), Vol. 24, No. 6, p. 332 (2pp)

Segmentation, relationships, individuals

Acknowledges the value of consumer segmentation as a means of targeting mass marketing efforts more effectively. But emphasises that bricks-and-mortar stores are not in fact dealing with segments, but with individual people. Takes an instance of a specific store selling electronic products, which has adopted the stance that, since its supposed market segment is young persons, the store should be filled at all times by overpoweringly loud noise, which is held to be an appropriate shopping environment for the target segment. Points out that this approach not only misunderstands the nature of a demographic segment, by no means all of whose members share the same characteristics, but also ignores the preferences of the members of all the other segments, numbers of whom may well be prevented from patronising this store — or patronising it more than once — even though it may stock goods relevant to their needs. Suggests that the propensity to treat people as though their needs/tastes could be predicted from their presumed membership of this or that demographic segment, is not far removed from the solecism of treating people according to the abundance of their hair or the colour of their skin.

A slight piece, which nevertheless makes a point which can hardly be made too often: however unavoidable segmentation may be in marketing, customer relationships are with people.

Research: - - - - - Practice:*** Originality:* Readability:****

Ref: 9312

The effects of corporate social responsibility and price on consumer responses

L.A. Mohr and D.J. Webb

SURVEY. The Journal of Consumer Affairs (UK), Vol. 39, No. 1, p. 121 (27pp)

Corporate social responsibility, CSR, stakeholders, shareholders

Notes increasing pressure (post-Enron) on companies to take on more community responsibilities. Claims that nearly 50 per cent of large corporations have programmes with a social issue, allegedly as a means of improving brand equity and corporate image. Notes the difficulty of reconciling these social demands with those of shareholders. Notes a debate in academia over whether companies should pursue a stakeholder theory (considering the interests of all constituencies) or shareholder theory; claims that most companies tilt towards their shareholders; moreover this is increasingly held to involve concentration on maximising (short-term) stock prices, whereas expenditure on corporate social responsibility (CSR) may involve a long wait for a pay-off. Notes fears that attempts to recoup CSR costs in higher prices may lead to lower sales. Suggests that in fact CSR may add value to a product. Describes, therefore, an experiment to examine whether 194 random US adults, in a role-playing situation, would be influenced in their propensity to buy, at variable prices, by the reported level of CSR (relating to either environmental or philanthropic activity) of the hypothetical retailer. Lists a series of hypotheses, and details how these were tested. Finds that, in relation to both environmental concerns and philanthropy, perceived CSR positively affected both evaluation of a company and purchase intent; moreover environmental CSR affected purchase intent more strongly than price.

The admitted unreality of the scenario created for this research means that quantified conclusions are impossible. But qualitatively the findings are interesting, and deserve the setting up of a much more rigorous test bed.

Research:*** Practice:* Originality:** Readability:***

Ref: 9313

Zopa: Web 2.0 meets retail banking

M. Kupp and J. Anderson

CASE STUDY. Business Strategy Review (UK), Vol. 18, No. 3, p. 11 (7pp)

Web 2.0, loans, freeformers, borrowing, Zopa

Examines the Web 2.0 phenomenon of Zopa, an internet start-up peer-to-peer brokerage business that brings together individual willing lenders and borrowers online. Indicates that, in two years to March 2007 Zopa had signed up 140,000 people. Shows that the aim is to bring together credit-worthy (A*–C only) borrowers (typically 'freeformers', eg self-employed or freelance workers not in standard full-time employment) with individual lenders. Prospective lenders set their own required interest rates, sums available and length of loans; borrowers state requirements; Zopa matches the two and borrowers select from loans on offer. Loans are spread between many lenders and repaid by monthly direct debit. Claims that borrowers get cheaper loans than available to them from most banks; lenders do better than from traditional savings accounts. Asks the question whether Zopa has created a sustainable model. Considers its physical resources, human resources and organisational resources, questioning to what extent these are valuable, rare and difficult to imitate or substitute. Judges that Zopa's resources could fairly easily be replicated by a determined competitor. Considers whether Zopa is strengthened by possession of first-mover advantage; notes that it has acquired a significant (but not large by internet standards) customer base, which might find it difficult or expensive to switch. Concludes that, while Zopa's resources are valuable and (pro tem) rare, its survival is far from secure — although the peer-to-peer lending model is here to stay.

A fascinating study of a win-win-win facility, such as only the Web could make possible.

Research:** Practice:** Originality:***** Readability:*****

Ref: 9314

How valuable is word of mouth?

V. Kumar, J.A. Peterson and R.P. Leone

THEORETICAL, with example. Harvard Business Review, October 2007, p. 139 (7pp)

Word of mouth, CLV, customer referral value, CRV

Claims that customer value resides not only in customer purchases, but also in referrals. Notes that most companies go no further than estimating customers' willingness to make referrals, whereas such good intentions are frequently not followed through. Quotes an actual case where 68 per cent of a financial services company's, and 81 per cent of a telecoms company's customers expressed willingness to refer, only 33 and 30 per cent respectively did — and of the referrals made, very few resulted in profitable new customers. Describes a generalised method for calculating a customer's lifetime value (CLV); claims that to this sum should be added a calculation of customer referral value (CRV). Gives a generalised formula for both forms of calculation, which are based on analysis of past behaviour, and projection forward. Calculates a CLV for a customer from each of the two companies surveyed; indicates that past referral behaviour suggests each will (after the launch of an incentivised referral programme) refer four new customers (of whom half are genuinely new, and half would have become customers anyway). Demonstrates that, when the CLVs of these referred customers are calculated, their combined value far outstrips the CLV of the original customer. Contends that customers with the highest CLVs are not those with the highest CRVs. Details a further campaign to the telecomm customers designed to improve referrals of high CLV customers, and CLV of high CRV customers.

An interesting concept, well-written and worth reading. But there's something about the rather slick mathematics that you may find unconvincing: real life is rarely (never?) as tidy as these examples make it seem. (And don't be put off by the unnecessarily complex-looking algebraic formulae for CLV and CRV.)

Research:** Practice:*** Originality:*** Readability:****

Ref: 9316

In the mood for advertising

F.E. Bonner, J.R. Bonner and J. Faase

EXPERIMENTAL. International Journal of Advertising (UK), Vol. 26, No. 3, p. 333 (23pp)

Advertising, Sundays, mood

Notes previous research findings that advertising is more effective if the recipient is in a happy, benevolent mood. Further, that studies in this field have all been in laboratory conditions of a non-realistic kind. Describes a field study conducted with 1,287 persons, to determine whether advertisement recall is affected by mood at time of seeing advertisement, and whether respondents were generally in a better mood on Sundays than at other times of the week. Gives details of how mood measurement (using eight criteria) and reading behaviour (three criteria) was measured. Advances five possible reasons why higher ad noting in Sunday papers may exist: size of paper; size of ad; content; creativity; reading time. Gives reasons for ruling out these explanations in the context of this survey. Concludes that mood is the main explanatory factor for higher advertising recall in relation to Sunday newspapers. Considers the implications of this finding for marketing: advertisers may choose to advertise in a Sunday paper rather than on another day; newspapers may choose to charge higher rates for space in a Sunday edition, based on higher recall values; others may find this an additional argument for considering a Sunday edition.

A fairly heavy piece of research to underpin some rather intuitively acceptable findings.

Research:*** Practice:** Originality:** Readability:**

Ref: 9318

Deterring illegal downloading: The effects of threat appeals, past behaviour, subjective norms and attributions of harm

A.M. Levin, M.C. Dato-on and C. Manolis

SURVEY. Journal of Consumer Behaviour (UK), Vol. 6, No. 2/3, p. 111 (12pp)

Music, downloads, illegal downloads

Describes two experiments conducted with 388 undergraduate students, to determine the effectiveness of a number of strategies on tendency illegally to download music from the internet. The first study measured the effect of a threat of low, medium or high level of potential sanctions, in conjunction with an attribution of harm caused to recording companies, or artists. The second study intensified the concept of harm caused to others, while also introducing questions of what subjects thought the social norms of approval/disapproval for this kind of activity were. Both studies were administered by pencil and paper. These studies found that stronger apparent threats did reduce the likelihood of future illegal downloading, while those who had indulged in this practice in the past were more likely to do so in the future. Attribution of harm to others had no effect on future intentions. However, social norms, as perceived by the subjects, did play a part in discouraging illegal behaviour — but only when the threat of sanctions is low, and the subject is not a habitually high downloader. Considers implications for marketers of electronic media.

The curious thing in this study is an apparent failure to examine the effect of the argument that illegal downloading may eventually lead to catastrophic consequences for the music industry as a whole—and hence for music lovers generally.

Research:** Practice:* Originality:** Readability:**

Ref: 9319

Editorial

J. Ozimek

EDITORIAL. Journal of Database Marketing & Customer Strategy Management (UK), Vol. 14, No. 4, p. 263 (3pp)

Marketing, analysis

Notes marketing's need for analysis — including complex analysis to deal with complex situations. Contrasts the simplicity of measuring response rates to, for example, a direct mail promotion, with the complexity involved in the contemporaneous use of several channels; notes the unwillingness of some marketing managers to accept the need for a complex model to explain a complex situation. Instances also the ways in which the occurrence of one positive phenomenon may have an unexpected effect on another, negative, event — for example increasing life expectancy giving rise to growing incidence of deaths from prostate cancer. Emphasises the demand in business for simplicity and straightforward messages; notes that academic theory advances in small incremental steps, which marketing can't afford to do. Claims that information and insight based on analysis should be cultivated — but in a balanced way, with intelligible communication between specialists and creatives.

A sane and balanced appeal to academia to consider practitioners' needs—and to practitioners not to reject or neglect what analysts and such-like specialists can provide.

Research: - - - - - Practice:*** Originality:** Readability:****

Ref: 9320

Lessons learned: A case study using data mining in the newspaper industry

C.L. Gunnarsson, M.M. Walker, V. Walatka and K. Swann

THEORETICAL, WITH CASE STUDY. Journal of Database Marketing & Customer Strategy Management (UK), Vol. 14, No. 4, p. 271 (10pp)

Data mining, data warehouse, decision trees

Defines data mining as an iterative process of knowledge discovery that promotes data-driven business intelligence and decision making. Holds that it is relatively new, hampered by privacy concerns, and limited by companies' functional silos. Defines a series of five steps in the process: Define business objectives; create data warehouse; prepare data (holds that this absorbs 85 per cent of data mining effort); modelling (distinguishes between data mining and statistical modelling by stressing that the former does not rely on assumptions and limitations); data-driven decision making. Notes the three most frequent modelling techniques — decision trees, regression analysis and neural networks. Describes the progress through the above five stages in relation to a data mining initiative in a US newspaper in the Midwest, with the objective of reducing churn. Notes the problems of three ill-matched files (on Households, Promotions and Transactions). Describes data cleansing, and the problems of free text format, missing data, duplicate records, contradictory data. Describes the use of a decision tree modelling process, and why this method was chosen. Urges standardisation of data mining through exchange of techniques within an industry.

The authors perhaps underestimate the true extent to which data mining is already practised—but this is a valuable introduction for those not yet familiar with the concept.

Research:** Practice:*** Originality:* Readability:***

Ref: 9321

Implementation study: Using decision tree induction to discover profitable locations to sell pet insurance for a startup company

R. D'Souza, M. Krasnodebski and A. Abrahams

CASE STUDY. Journal of Database Marketing & Customer Strategy Management (UK), Vol. 14, No. 4, p. 281 (8pp)

Data mining, genetic algorithms

Notes that the search for data insight has moved from the use of OLAP tools towards data mining. Describes five categories of data mining tasks: Dependency Analysis; Class Identification; Concept Description; Deviation Detection; Data Visualisation. Focuses on Class Identification. Identifies two algorithms used in this study (SBP and C4.5). Outlines the data collection phase (census data, and vet location data), and the running of data mining software on a training set of 2,000 records: gives the results in the form of generated rules, which were then validated on three further test data sets of 2,000 records each. Indicates the profit predicted for each test set by applying these rules. Shows a lift chart, indicating the lift in response, decile by decile, for each of the two algorithms employed, and a gains chart for each algorithm showing similarly the lift in profit, compared with random selection of prospects. Suggests the use of genetic algorithms to combine the two algorithms used in this study for still better results.

Not the best illustration of a gains chart, since this one indicates that, although 30 per cent of prospects generate 80 per cent of profit, maximum profit (as well as maximum response) would be achieved by promoting 100 per cent of prospects.

Research:*** Practice:*** Originality:** Readability:**

Ref: 9322