can loyalty schemes really build loyalty?

[ 47 ]

Marketing Intelligence &Planning16/1 [1998] 47–55

© MCB University Press [ISSN 0263-4503]

Can loyalty schemes really build loyalty?

Lisa O’MalleyCardiff Business School, University of Wales, College of Cardiff, UK

Customer loyalty schemeshave blossomed in the era ofcustomer retention, and havebeen willingly embraced byboth retailers and consumersalike. Today’s loyalty schemesare modelled on the AAdvan-tage Programme; a frequentflier programme initiated byAmerican Airlines in 1981,and lessons from the airlineindustry are briefly intro-duced. The primary purposeof this paper is to evaluatethe extent to which loyaltyschemes really can buildloyalty. In order to do this,four categories of loyalty asidentified in the literature areconsidered. Given existingcriticisms of loyalty schemesit is concluded that suchschemes have an importantrole to play in situationswhere no loyalty or spuriousloyalty is evident. However,where sustainable loyalty isthe ultimate goal, customerloyalty schemes are of impor-tance only as part of a coher-ent value proposition.

Introduction

Loyalty schemes (LSs) are now perceived tobe one of the most successful marketing toolsof the 1990s, and have received considerableattention in both trade and academic jour-nals. The customer loyalty scheme is oneexample of the continual blurring of theboundaries between direct marketing and“mainstream” marketing, relying as it doesupon a customer database and direct commu-nications. It has been suggested that the cus-tomer loyalty scheme arena is where today’smarketing battles are being fought and,indeed, the plethora of LSs available to con-sumers certainly suggests that this may betrue. Both retailers and consumers appear tohave willingly embraced customer loyaltyschemes, albeit for different reasons. Giventhe apparent success of the customer loyaltyphenomenon it seems timely to take a closerlook. The purpose of this paper is therefore:1 To review the motivation for the develop-

ment of customer LSs.2 To discuss the concepts of “loyalty” and

“value”.3 To identify the most common objectives of

customer LSs.4 To evaluate critically customer LSs, par-

ticularly in terms of the extent to whichloyalty can be encouraged.

Recognition of the importanceof customer retention

Customer retention has always been impor-tant, although traditionally it has beenviewed rather simply as an outcome of suc-cessful marketing. This is embodied in thedescription of marketing as “selling productsthat do not come back to customers that do”!Because retention was seen as a by-product ofmarketing, organisations focused on cus-tomer acquisition with the explicit aim ofincreasing market share. This, in turn, wasassumed to increase both the organisation’spower and profitability. As long as marketshare continued to grow, it was accepted thatmarketers had got it right. However, market

share, as an aggregate measure, can cover upa host of inadequacies in marketing terms(Grönroos, 1994), not least of which is the factthat increases in market share disguise thelevel of customer defection. Rosenberg andCzepiel (1984, p. 46) observe that “some com-panies seem hooked on steady doses of freshcustomers to cover up regular losses of exist-ing ones”. It was not until the early 1990s thatthere was widespread recognition that thisrepresented a serious flaw in marketing.Reichheld and Sasser (1990) calculated theimpact of customer retention on profitability:“As a customer’s relationship with the com-pany lengthens, profits rise. And not just alittle. Companies can boost profits by almost100 per cent by retaining just 5 per cent moreof their customers” (Reichheld and Sasser,1990, p. 105). Reduced to pounds, shillings andpence in this way the business world began tofocus aggressively on customer retention.The aim was to reduce marketing’s “scrapheap” (customers who defect) and, thus,enhance profitability. This renewed focus oncustomer retention is generally referred to asloyalty marketing, and is increasingly relianton a customer database and direct communi-cation.

LSs have blossomed in the era of “customerretention” though their emergence is not arecent phenomenon. They have been aroundin one form or another since the early 1970s.The basic idea of a loyalty programme is toreward customers’ repeat purchasing andencourage loyalty by providing targets atwhich various benefits can be achieved. Anearly example of a Frequent FlierProgramme (FFP), Southwest Airlines“Sweetheart Stamps” promotion in the 1970s,allowed business travellers to accumulatebenefits in order to take their partners on afree flight (Gilbert, 1996). In the UK, GreenShield Stamps promotions provided similarincentives. These schemes were reasonablysuccessful, despite their lack of sophistica-tion. What distinguishes today’s schemes istheir ease of use for consumers, and thedegree of sophisticated targeting made possi-ble through continual improvements in tech-nology.

[ 48 ]

Lisa O’MalleyCan loyalty schemes reallybuild loyalty?


Origins of loyalty schemes

Although perceived as a descendent of GreenShield Stamps promotions, today’s LSs moreclosely resemble early Frequent Flier Pro-grammes (FFPs). Indeed, the majority ofcustomer loyalty programmes in operationtoday are modelled on the “AAdvantage Pro-gramme” introduced by American Airlines in1981. In the wake of industry de-regulation in1979, American Airlines were looking for acost-effective marketing proposition whichcould fulfil the dual objectives of promotingcustomer loyalty and providing consistencyof demand. One solution was to transformotherwise underutilised capacity into a mar-keting tool for rewarding loyalty (Gilbert,1996). American Airlines trawled its Cus-tomer Reservation System (CRS) “SABRE” inorder to identify the names of Frequent Fliersso that they could initiate their programme.The FFP allowed passengers to accumulateair miles in direct proportion to their use ofthe airline, and was particularly attractive tobusiness travellers who were already flying agreat deal, and who could accrue benefits tobe used either in terms of up-grades, or asfree flights in their spare time.

The “AAdvantage Programme” was quicklycopied by other airlines, hotels and evencredit cards in the USA (O’Brien and Jones,1995). Similar programmes became popularin Europe by the early 1990s, and have sincebeen developed in Asia and the Far East(Gilbert, 1996). On the one hand, the phenom-enal up-take of LSs suggests that they must beeffective. Alternatively, however, there is theargument that loyalty programmes are seri-ously over-rated, that is, despite the hyper-bole surrounding them, it seems that all FFPsand guest LSs have not been successful, andhave often proved too expensive to implement(Gilpin, 1996). Indeed, although FFPs appearto represent a very powerful tool for airlines,there are a number of potential problemswhich should be a warning for similarschemes (Gilbert, 1996). These include corpo-rate hostility towards FFPs; governmentthreats to tax FFPs; the potential impact ofunredeemed miles; and the proliferation of“me too” schemes. The first three of these aredirectly related to the airline industry, whilethe final problem is more universal, in thatthe “me too” criticism is clearly problematicfor all industries where loyalty programmesare in operation. That is, “the failure to con-sider that when everyone offers similarrewards there is no competitive advantageand the cost of retaining existing, or winningnew customers, may increase” (Gilbert, 1996,p. 138).

In terms of this last problem, Gilbert out-lines that the growth in use of FFPs hasclearly affected airlines’ profitability andeffectiveness over the last ten years. However,individual airlines are reluctant to abandontheir schemes, as this will result in a competi-tive disadvantage. As a result, “the schemesare here to stay as … once FFPs are estab-lished, it is extremely difficult, if not impossi-ble to discontinue them” (Gilbert, 1996, p. 144).

Despite the apparent problems associatedwith airlines’ FFPs, the retail loyalty schemehas been widely embraced here in the UK, bypetrol retailers, supermarkets, bookstoresand a host of other organisations. Dowlingand Uncles (1997, p. 74) attribute the speed ofLSs adoption to competitive factors. “Onceone retailer broke ranks, all others followedwithin months”. That said, the majority ofretailers would argue that LSs are ultimatelymore about engendering loyalty than match-ing competition. As such, it is timely toexplore exactly what is understood by theconcept of loyalty.

What is loyalty and how is loyaltyincreased?

Attempting to define loyalty is far easier saidthan done. The problem lies in identifyingwhether loyalty is an attitudinal or behav-ioural measure. “Used loosely, as it usually is,the term “loyalty” conjures up variousnotions of affection, fidelity or commitment”(McGoldrick and Andre, 1997, p. 74). This hasled to the use of customer satisfaction as aproxy measure of loyalty because it has beenassumed that satisfaction affects buyingintentions in a positive way. However,research indicates that it is overly simplisticto assume that dissatisfied customers willdefect, and that satisfied customers willremain loyal. Indeed, Reichheld (1994) foundthat despite being “satisfied” or “very satis-fied” many customers still defect. In the UK,Oglivy Loyalty Centre found that, although 85per cent of automotive customers reportedbeing satisfied, only 40 per cent repurchased,and 66 per cent of packaged goods customerswho identified a favourite brand had bought“another brand” most recently (McKenzie,1995). Such behaviour may be explained bythe impact of other variables such as choice,convenience, price, and income. This sug-gests two things; first, attitudinal measures ofsatisfaction are poor predictors or measuresof behaviour, and second, it casts some doubton the concept of 100 per cent loyalty. Indeed,it has been suggested that few customers are100 per cent loyal to a single brand, but rather,are likely to have a repertoire of two or three

[ 49 ]



brands within any product category fromwhich they regularly buy – i.e. polygamousloyalty (Dowling and Uncles, 1997).

In recognition of the limitations of satisfac-tion as a proxy measure for loyalty, behav-ioural measures such as recency, frequencyand monetary value (typical measuresutilised in direct marketing) are increasinglybeing used. These provide a more realisticpicture of how well the brand is doing vìs-à-vìs competitors, and the data generatedfacilitate calculation of customer life-timevalue, enhance prediction of purchase proba-bilities, and assist in developing cost-effectivepromotions. Greater use of behavioural mea-sures has been made possible by advances indatabase and tracking technologies.

However, despite the ability to generate amore accurate picture of where and howmuch customers are spending, it has beenargued that “loyalty implies a commitment toa brand that may not be reflected by just mea-suring continuous behaviour” (Assael, 1992,p. 87). In particular, behavioural measures arelimited in determining the factors whichinfluence repeat purchase (Dick and Basu,1994). “Behavioural measures are insufficientto explain how and why brand loyalty isdeveloped and/or modified… high repeatpurchase may reflect situational constraints,such as brands stocked by retailers, whereaslow repeat purchase may simply indicatedifferent usage situations, variety seeking, orlack of brand preference within a buyingunit” (Dick and Basu, 1994, p. 100). Althoughbehavioural measures do overcome some ofthe limitations of using satisfaction as aproxy measure for loyalty, it is clear thatbehaviour is also insufficient as a measure ofstore or brand loyalty (Dick and Basu, 1994;Knox, 1996; Knox and Walker, 1995;McGoldrick and Andre, 1977; Solomon, 1996).As a result, the extent to which behaviouralmeasures are of benefit in developing long-term loyalty strategies is clearly question-able.

Therefore, it seems that neither attitudinalnor behavioural measures on their own aresufficient to explain or define loyalty. This isimportant for managers, as it suggests thatexisting measures of loyalty may be seriouslyflawed, and thus strategies developed on thestrength of such measures may beinadequate. “The very term loyalty impliescommitment rather than just repetitivebehaviour, which suggests that there is a needfor a cognitive as well as a behavioural view”(Assael, 1992, p. 89). A number of frameworkshave been developed which combine bothattitudinal and behavioural measures ofloyalty in an effort to overcome such prob-lems (Dick and Basu, 1994; Knox, 1996;

McGoldrick and Andre, 1977). The frameworkproposed by Dick and Basu (1994) is particu-larly useful because not only does it offermore accurate measures of loyalty, but it alsoresults in a number of different categories ofloyalty, each of which have clear managerialimplications (see Table I). They utilise twodimensions or measures of loyalty:

Relative attitude: Relative attitude not onlyfocuses on attitude to the entity, but alsoincorporates attitudes to alternatives. Thisencapsulates not only satisfaction measures,but also preference measures. The suggestionis that the higher the relative attitudebetween alternatives, the more likely thatattitude will influence behaviour.

Patronage behaviour: This includes tradi-tional measures such as recency, frequencyand monetary (RFM) value of purchases,share of wallet, purchase sequence, etc.

Using Dick and Basu’s (1994) conceptualisa-tion of loyalty, it is possible to determine theprobable role of LSs in engendering loyalty.LSs have a particularly important role to playin situations where there is either no loyaltyor spurious loyalty, but that role is dimin-ished considerably in situations where latentor sustainable loyalty is displayed.

Petrol retailers have had to contend withthe “no loyalty” situation. For example,research carried out by Shell in the early1990s suggested that 85 per cent of petrol buy-ers were not loyal to any brand or location(Dignam, 1996). Competing offers are gener-ally perceived as being undifferentiated, andpetrol retailers have attempted to increasecustomers’ switching costs in order toincrease retention. One effective method ofachieving this is through “point accrual pro-grammes” as used within many existing LSs(e.g. Tiger Tokens, and Premier Points).These programmes aim to tie the customerinto a particular supplier through thepromise of future rewards, where the pointsalready accrued represent switching costs.Thus, in no loyalty situations, managers canattempt to generate spurious loyalty throughsuch means as in-store promotions, loyaltyclubs and special offers (Dick and Basu, 1994).

Spurious loyalty is very similar to the con-cept of inertia, where despite perceptionsthat choices are relatively undifferentiatedbehavioural data suggest loyalty. In suchcases, repeat purchase may be based on theavailability of deals, special offers, conve-nience or the influence of other people. As aresult, the consumer may only temporarilydisplay such loyalty, and is likely to be veryopen to competing offers (i.e. promiscuous).That is “if another product comes along thatis for some reason easier to buy (e.g. it ischeaper or the original product is out of

[ 50 ]



stock), the consumer will not hesitate to doso” (Solomon, 1996, p. 290). As a result it isconsidered relatively easy to encourageswitching behaviour in markets typified byspurious loyalty. In order to defend the cus-tomer base against competitive attacks man-agers can attempt to maintain spurious loy-alty and/or develop sustainable loyalty. How-ever, one problem associated with simplymaintaining spurious loyalty is that the com-pany may become involved in a continualspiral of promotions, because, once the incen-tive is taken away, customers have lost theirprimary reason for purchase (Dowling andUncles, 1997).

Latent loyalty occurs when a consumer hasa high relative attitude towards the companyor brand, but this is not evident in terms oftheir purchase behaviour. This is probably asa result of situational influences – includinginconvenient store locations, out-of-stocksituations, and/or the influence of other peo-ple. In the case of latent loyalty, Dick andBasu (1994) outline that managerial effortsare best focused on removing the obstacles topatronage, for example by extending thebranch network. The only situation wherethe problem of latent loyalty may be directlyaddressed by an LS is where a customer hasdefected because of the existence of a compet-ing scheme.

Sustainable loyalty (O’Brien and Jones,1995) exists when the customer exhibits highrepeat purchase, and does so because theyhave a strong preference (high relative atti-tude). Sustainable loyalty is thereforeachieved “when the company has developedand communicated a proposition that clearlyhas long-term benefits for the customer”(O’Brien and Jones, 1995, p. 97), and where thecustomer modifies his or her behaviour to

remain loyal over time. Thus sustainableloyalty occurs where repeat patronage isaccompanied by a favourable attitude(Aasael, 1992; Dick and Basu, 1994), that iswhere purchase is as a result of a consciousdecision by the consumer (Solomon, 1996). Assuch, this is clearly the most preferred of thefour categories, and may be what we intu-itively mean by loyalty. Organisations mustnot become complacent, however, and, man-agerial efforts should be continually focusedon reinforcing attitude and behaviour, partic-ularly as the consumer is likely to be targetedby competitors. This may involve maintain-ing a price advantage, and/or providing addi-tional services which offer value to the cus-tomer.

Loyalty schemes and the concept of valueThere is an assumption that LSs providebenefits which represent “value” tocustomers, and it is because of this that LSscan encourage loyalty. However, the extent towhich LSs offer “value” to consumers is alsoquestionable, particularly because value willrepresent different things to different people,and will be different in different contexts.Nevertheless, O’Brien and Jones (1995, pp. 79-80) identify five elements which determinethe value of an LS. These include cash value(how much the reward represents as a pro-portion of spend); choice of redemptionoptions (the range of rewards offered); aspira-tional value (how much the customer wantsthe reward); relevance (the extent to whichrewards are achievable); and convenience(ease of participation in the scheme).

Retail LSs in the UK typically offer a rebateequivalent to 1 per cent of purchases (cashvalue), with rewards redeemable either in the

Table ICategories of loyalty

Category No loyalty Spurious loyalty Latent loyalty Sustainable loyalty

Relative attitude Low Low High HighPatronage behaviour Low repeat High repeat Low repeat High repeatManifestation Does not patronise the Patronises the company, but In this case, the customer The individual enjoys a

company, and does not wish to does not have a high relative wishes to patronise the high relative attitude, attitude. This may be as a result organisation, but perhaps is together with high repeat of other factors, including not able to do so – store patronage behaviourlocation convenience, lack of location may be inconvenient,alternatives favourite brands not stocked,

etc.Implications Management may “Spurious loyalty” cannot be Managerial efforts are best Loyalty must be

attempt to generate relied on. The customer is focused on removing the continually reinforced, “spurious loyalty” clearly open to better offers obstacles to patronage for and the value offered

the customer must remain acceptable

Source: Adapted from Dick and Basu (1994)

[ 51 ]



form of discounts/vouchers and/or restaurant/hotel discounts and air miles(redemption of options value). While cus-tomers are unlikely to aspire actively to astraight discount on purchases, holiday or airtravel rewards are of a higher aspirationalvalue. Furthermore schemes are more rele-vant when customers are capable of achiev-ing the rewards they value. For example,having to spend over £2,000 on groceries toearn a free trip is hardly relevant toconsumers when they calculate the relativecosts involved. Finally, since most schemestoday are card-based, they are far more conve-nient for customers than earlier sales promo-tions because they do not have to worry aboutretaining receipts, etc. Although it isacknowledged that few schemes today offerall five elements of value, it is clear that “com-panies who want to play the rewards gameshould be sure that their value measures upto customers’ alternatives” (O’Brien andJones, 1997, p. 79). This is particularly impor-tant where the loyalty scheme is the mainpoint of differentiation, as may be true in thecase of spurious loyalty.

Objectives of loyalty programmes

Loyalty programmes are developed for avariety of reasons including to reward loyalcustomers, to generate information, tomanipulate consumer behaviour, and as adefensive measure to combat a competingscheme.

Reward loyal customersMost retailers implicitly suggest that the aimof their scheme is to reward repeat purchas-ing. This is achieved through a combinationof discounts and other “rewards”. Consumersare motivated to participate in such schemesbecause, fundamentally, most people like toget something for nothing (Uncles, 1994). Thisargument is supported by recent empiricalresearch, where participants in group discus-sions agreed that, if a customer already patro-nises a particular store, then it makes senseto join the scheme (Evans et al., 1997).

To generate information“Knowing who the best customers are, whatthey buy, and how often provides a secretweapon” (Stone, 1994, p. 37). However, untilrecently, store-based retailers have not beenin a position to recognise their customers, letalone identify their best customers. Unlikedirect marketers, airlines, hotels or evenbanks, store-based retailers did not automati-cally collect and record individual customers’details. Indeed, most retail systems

(including electronic point of sale) have beendesigned to capture information on productsnot customers. Improvements in databasecapacity provided part of the solution; how-ever, retailers had to contend with an addi-tional problem in terms of how to justify andorganise data collection. LSs provided retail-ers with both the justification and mecha-nism to collect personal data. Because of therewards offered, customers have willingly co-operated in terms of data provision, and arehappy (it seems) for retailers to collect detailsof all their purchasing behaviour. As a result,retailers have been able to combine merchan-dise and customer profile information. Theinformation held on individual customerscan also be linked with data from a variety ofother sources (geodemographics, lifestyleand credit history) in order to build up afairly accurate picture of a customer’s life.Tactically, this information can be used toimprove targeting, create offers and shiftmerchandise. Strategically, holders ofdetailed customer information have a valu-able springboard to diversify into areas suchas financial services, as evident in Sains-bury’s recent move.

Manipulate consumer behaviourAlthough the stated aim of most schemes is toreward loyal customers, the fundamental aimof most schemes is to manipulate consumerbehaviour within a sophisticated system,where incentives and coupons can be individ-ually targeted, in order to encourage cus-tomers to try new products or brands;increase multi-pack purchases; pay premiumprices, and/or use the brand for increasinglydiverse services. In terms of the latter, somesupermarkets have successfully persuadedcustomers to purchase pharmaceuticals,petrol, wine and more recently even bankingservices. Thus, one of the most successfulapplications of the loyalty scheme may be interms of reducing the risk inherent in diver-sifying.

As a defensive measure to combat acompeting schemeIn practice the “decision to launch aprogramme is often motivated as much byfears of competitive parity as anything else,which companies rarely state publicly”(Dowling and Uncles, 1997, p. 73). This is evi-dent in the speed of uptake of FFPs in theearly 1980s following the launch of the “AAd-vantage Programme” and the number ofsupermarket schemes launched in the wakeof Tesco’s “Clubcard” in 1995. One of the prob-lems with LSs as a defensive move is thatorganisations are often unclear about whatexactly the scheme is intended to achieve. As

[ 52 ]



a result “reward programmes are widelymisunderstood and often misapplied”(O’Brien and Jones, 1995, p. 75).

This suggests that there is too much empha-sis on short-term results, with strategic objec-tives and implications often overlooked. Thatis not to say that there are no successful loy-alty programmes, but that loyaltyprogrammes per se may not necessarily bethe direct marketing panacea of the 1990s.Indeed, LSs are increasingly attracting criti-cism, including: that they are little more thansophisticated sales promotions; that loyalty isexhibited to the programme not the brand;that data collection is over-emphasised; andthat subsequently the costs may outweighadvantages.

They are little more than sophisticatedsales promotionThe origins of today’s retail LSs and theircontinued emphasis on discounts suggestthat they are not that far removed from theGreen Shield stamps promotions of the 1970s.Indeed, “too many companies treat rewardsas short-term promotional give-aways orspecials of the month” (O’Brien and Jones,1997, p. 75). The emphasis on discounting hasbeen likened to the “commercial equivalentof cocaine snorting – where an illusory short-term benefit leads to long-term disaster”(Bird, 1991, p. 27). This is a worrying criti-cism, because clearly the potential for moreeffective marketing is made possible by theuse of sophisticated databases. “In order for arewards programme to be a profit centreinstead of a cost centre, the payout must beinextricably linked to the desired (consumer)behaviours” (O’Brien and Jones, 1995, p. 77).This requires a thorough understanding ofcustomers’ expectations, and effective use ofthe information generated by the programmein order to achieve scheme objectives. So far,few organisations have utilised their LSs inthis way.

Although many schemes offer more thanjust a straight discount on purchases, cus-tomers have become somewhat jaded by theavailable “rewards”. For example, in theUSA, where LSs have a much longer history,“customers are so inured to offers promisingeverything from a free vacation to Florida toa free credit card that they either yawn whenthey see a new one or become experts at get-ting something for nothing” (O’Brien andJones, 1997, p. 77). This suggests that com-petitors may become involved in a constantbattle for one-upmanship, simultaneouslyincreasing customers’ expectations and thecosts associated with running theprogramme.

Loyalty to the scheme not the brandSales promotions have been criticised forencouraging promiscuity among customers,where customers shop for the best “deal”rather than being loyal to any one brand. It isclear that LSs may also suffer from this prob-lem, where too much emphasis on promotingthe scheme to consumers results in a shift inloyalty from the company or brand to loyaltytowards the scheme (Dowling and Uncles,1997; McKenzie, 1995). Although LSs maycreate an initial point of differentiationbetween brands, they are difficult to sustainas a prime reason for purchase, because fun-damentally “… it is difficult to increase brandloyalty above the market norms with an easy-to-replicate ‘add on’ like a customer loyaltyprogramme” (Dowling and Uncles, 1997, p. 94).

Schemes over-emphasise data collectionAlthough the ability to collect information isstressed as one of the principal benefits ofcard-based LSs, there are now suggestionsthat this may be misguided. As Mazur (1997,p. 16) states, “sitting here in 1997, …one couldbe forgiven for wondering whether, in theirdesperate struggle to figure out who theircustomers are and how to keep them, toomany marketers have been jumping on thecustomer loyalty bandwagon”. Although thecollection of customer data is clearly animportant element, too many schemes over-emphasise data gathering, to the extent thatthey now suffer data overload. As a result,there are often very basic uses of the informa-tion, with little evidence as yet of the poten-tial targetability of individual offers. Further-more, the information gathered through anLS may be insufficient as an input into strate-gic decision making. As Dowling and Uncles(1997, p. 73) identify, “such a self-selectedgroup is unlikely to represent all a company’spotential customers. Hence it is only onesource of market research information”.Indeed, it may be that too many companieshave emphasised database building overloyalty building, and this may have seriousimplications in the future.

Costs may outweigh advantagesThere is no doubt that for some sectors, andfor some companies LSs represent a cost-effective approach to marketing. However,because many organisations have been forcedinto offering a loyalty programme as a resultof competitive actions “... many customerloyalty programmes seem to have beenadopted too quickly and without muchthought” (Dowling and Uncles, 1997, p. 81). Aswith the airlines, when competitors offer thesame or similar rewards, there remains little

[ 53 ]



sustainable advantage. The problem is thatLSs are simply too easy to copy: although“evidence that the expenditure on loyaltyprogrammes is becoming a zero-sum game isstill anecdotal, … [it is] nonetheless indica-tive” (Mazur, 1997, p. 16). LSs are expensive toestablish and, other than the initial gatheringof data, are likely to take two to three years togenerate a return (Hochman, 1992). To putthis in perspective Shell is reported to havespent between £20 and £40m developing itsSmart Card scheme, and Tesco’s initial datagathering cost in the region of £10m (Dignam,1996; Reed, 1995). Given these substantialcosts LSs should clearly be viewed in terms ofa long-term strategic investment, rather thanin terms of short-term tactical manoeuvres.

Organisations have bought into LSsbecause they have accepted the economics ofcustomer retention. That is, managers haveaccepted that loyal customers are more prof-itable; loyal customers cost less to serve; loyalcustomers are less price-sensitive and; loyalcustomers generate positive word of mouth.Dowling and Uncles, in a review of the behav-ioural and loyalty literature, cast seriousdoubts on these assumptions: “in short, thecontention that loyal customers are alwaysmore profitable is a gross oversimplification.Every company needs to use its customerdata to determine the truth of this assertion”(Dowling and Uncles, 1997, p. 78). In any case,most organisations operate loyaltyprogrammes in addition to, not instead of,other marketing activities. As a result, thereis an increase in marketing expenditure.

How well do loyalty schemes manage loyalty?

On the basis of the discussion so far it seemsimportant that we re-evaluate the extent towhich LSs can actually manage customerloyalty. There seem to be two fundamentalproblems which limit the extent to whichlong-term sustainable loyalty can beachieved. These include the measurementproblems associated with loyalty, and thepredominant emphasis on schemes as short-term tactical, or defensive measures.

Measurement problemsOn the basis of the preceding discussion it isclear there are a number of issues whichmust be resolved if managers are to be in aposition to manage loyalty. First and foremostthere is a need to establish exactly what ismeant by loyalty, and to identify how it can bemeasured. As Reichheld (1994, p. 15) outlines,“if you can’t measure it – you can’t manageit”. Research suggests that a 100 per cent

loyalty is difficult, if not impossible toachieve, and that polygamous loyalty is farmore common. Indeed this is already evidentin customers’ multiple membership in loy-alty schemes (Dowling and Uncles, 1997), withUK consumers holding, on average 3.5 cardsin 1995 (Reed, 1995). If polygamous loyalty isall that we can hope for, then striving for 100per cent repeat purchase is an idealistic andexpensive aim. In terms of the conceptualissues, organisations need to determine whatcategory of loyalty they wish to achieve.Although sustainable loyalty is implicitlysuggested as the aim of most schemes, thecharacteristics of the scheme, and of manymarkets, inhibit its ultimate realisation (Dickand Basu, 1994). Indeed, in many low-involve-ment product categories developing andmaintaining spurious loyalty may be the onlyviable strategy. In such cases LSs may be acost-effective option, because they provide aninitial point of differentiation and incorpo-rate a “club” concept as a form of social influ-ence. However, where LSs are used to main-tain spurious loyalty, there is a continualneed to ensure that they provide greatervalue for customers than competing schemes(O’Brien and Jones, 1995). For example, Shellhave recently attempted to enhance “Redemp-tion of Options Value” by partnering with awider consortium of retailers, includingDixons, Curry’s Victoria Wine, VisionExpress, John Menzies, the RAC and HiltonHotels. Sustainable loyalty is clearly a morestrategic issue and, given the discussion sofar, it seems that LSs, as they are currentlyimplemented, are unlikely to fulfil this aim.“If they remain the only strategy, short-termincentives with little real effect on the bottomline will be the result” (Mazur, 1997, p. 16).

Tactical not strategic orientationAlthough card-based LSs offer the potentialfor high targetability, such use of customerdata is not yet widely evident. However, asorganisations become more adept at usingthe data they have, this situation may change.Unless data are effectively utilised, maintain-ing a database will become too costly, and thebenefits of superior targetability may neverbe realised. Equally, collection and use ofcustomer data within an LS is potentiallyintrusive (Evans et al, 1997) and, unless con-sumers see tangible benefits of data surren-der, they may become more protective of theirpersonal data (Patterson et al., 1997). A secondreason for an overly tactical emphasis may bethat the majority of schemes have been initi-ated as a defensive response to competitiveactivity. “This lemming-like rush to loyaltycards seems to indicate that the marketingfraternity is still offering too many tactical

[ 54 ]



promotions rather than focusing on theharder task of long-term strategic brandbuilding” (Mazur, 1997, p. 16).

Tactical issues may also be emphasisedbecause of the measures used to evaluatesuccess. These tend to be fairly short-term,and motivated as much by shareholder inter-ests as anything else. Therefore, managershave no choice but to “hit the numbers” inthe short term, limiting their emphasis onlong-term strategic issues.

LSs have been criticised for being ill-thought out, and badly implemented. How-ever, despite such criticisms it seems that LSsare here to stay. Whether they represent valuefor customers is clearly debatable, but cus-tomers are likely to continue to buy into this“something for nothing” proposition. It isequally debatable whether LSs offer value forretailers, but as with FFPs they have nowbecome a necessary requirement of doingbusiness in most retail sectors. As a result,retailers need to evaluate carefully their LSsand identify how they can be more effectivelyutilised. One important issue here is theresolution of the measurement problemsassociated with the concept of loyalty.

Marketers need to clarify the objectives oftheir loyalty programmes. They need todetermine the relative importance of datacollection, rewarding loyal customers, manip-ulating behaviour, erecting switching costs,and shifting merchandise, etc. In particular,they must assess the extent to which theirscheme(s) can, and do, influence sustainableloyalty. While behavioural measures of loy-alty (as generated by customer LSs) are ofobvious benefit, they tell only half the story.Therefore, reliance on behavioural measureswill have “obvious dangers for strategy for-mulation” (McGoldrick and Andre, 1977, p. 75). Consequently, retailers should continueto use other measures of loyalty, includingsatisfaction and relative attitude, and shouldnot rely solely on self-selected customer LSmembers.

Retailers also need to focus on the conceptof “value”, not just in terms of the valueinherent in the scheme, but also in terms ofthe total value offered to customers. When theLS becomes a component of the retail offer-ing, it will take its place alongside other ele-ments of value such as merchandise range,price, customer service, design, location andconvenience. This more holistic view of valuewill go much further in engendering sustain-able loyalty, than could be achieved by any LSon its own. Organisations which focus theirattention on developing appealing valuepropositions will be most successful in thelong term. At the very least, retailers must

remember that “loyalty has to be earned, itcannot be given away” (Mazur, 1997, p. 16).

Retailers are not currently making the bestuse of the information they hold. Indeed,despite emphasising the data collection bene-fits, too many LSs remain as simple “pointaccrual programmes”. Retailers can learn agreat deal from direct marketers in terms ofsophisticated data analysis, but this must beused to develop and communicate appealingoffers to individual customers. “Rewardsprogrammes do not exist in a vacuum; theymust dovetail with a company’s overall strat-egy and capabilities” (O’Brien and Jones,1995, p. 79). In order to make more effectiveuse of existing LSs, retailers would do well toconsider the following:1 Re-evaluate and clarify the objectives of

the programme.2 Be realistic as to the extent to which spuri-

ous or sustainable loyalty can be achieved,and allocate budgets accordingly.

3 Identify the relative importance of datacollection, customer reward, merchandisepromotion, etc. within the loyalty scheme.

4 Analyse and utilise the data generated bythe loyalty programme.

5 Supplement behavioural data analysiswith traditional research methods, andinclude non-card holders within this sam-ple.

6 Engage in direct marketing (offers andcommunication) which are relevant toindividual customers.

7 Be responsive to individual customers,and continually re-evaluate the “value”offered by the scheme (especially in com-parison with competing schemes).

8 Ensure that the loyalty scheme is a coher-ent element of the company’s overall strat-egy.

Conclusion

Many organisations have clearly benefitedfrom the implementation of a customer loy-alty scheme. However, for many others theyhave become a necessary and costly require-ment for doing business. As Dowling andUncles (1997, p. 71) suggest, “most schemes donot fundamentally alter market structure.They might help to protect incumbents andmight be regarded as a legitimate part of themarketer’s armoury, but at the cost of increas-ing marketing expenditures”. Although LSsmay have a valuable role in retail marketingtoday, they can achieve little more than spuri-ous loyalty when they are viewed simply as“point accrual programmes”. Such an empha-sis results in a continual game of one-upman-ship which is increasingly costly. Equally,where schemes are viewed simply as data

[ 55 ]



collection mechanisms, the purpose of thescheme tends to be obscured in a continualsearch for more and more detailed informa-tion. However, when schemes are consideredto be just one element of a coherent valueproposition, they begin to play an integralrole in developing sustainable loyalty. Thisseems to be the only viable, long-term role forcustomer loyalty schemes.

ReferencesAasael (1992), Consumer Behaviour and Marketing

Action, 4th edition, PWS-Kent, Boston, MA.Bird, D. (1991), “Don’t let them get away”, Market-

ing Direct, December, pp. 65-71.Dick, A.S. and Basu, K. (1994), “Customer loyalty:

toward an integrated framework”, Journal ofthe Academy of Marketing Science, Vol. 22 No. 2, pp. 99-113.

Dignam, C. (1996), “Being smart is not the onlyredeeming feature”, Marketing Direct, Sep-tember, pp. 51-6.

Dowling, G.R. and Uncles, M. (1997), “Do customerloyalty programmes really work?”, SloanManagement Review, Summer, pp. 71-82.

Evans, M.J., Patterson, M. and O’Malley, L. (1997),“Consumer reactions to database-basedsupermarket loyalty programmes”, The Jour-nal of Database Marketing, Vol. 4 No. 4, pp. 307-20.

Gilbert, D. (1996), “Airlines”, in Buttle, F. (Ed.),Relationship Marketing: Theory and Practice,Paul Chapman Publishing Ltd, London, pp.131-44.

Gilpin, S.C. (1996), “Hospitality”, in Buttle, F.(Ed.), Relationship Marketing: Theory andPractice, Paul Chapman Publishing Ltd, London, pp. 145-58.

Grönroos, C. (1994), “From marketing mix torelationship marketing: towards a paradigmshift in marketing”, Management Decision,Vol. 32 No.2, pp. 4-20.

Hochman, K. (1992), “Customer loyaltyprogrammes”, in Nash, E.L. (Ed.), The DirectMarketing Handbook, McGraw-Hill, NewYork, NY, pp. 781-99.

Knox, S. (1996), “The death of brand deference:can brand management stop the rot?”, Mar-keting Intelligence & Planning, Vol. 14 No. 7,pp. 35-9.

Knox, S. and Walker, D. (1995), “Empirical devel-opments in the measurement of involvement,brand loyalty and their structural relation-ships in grocery markets”, Cranfield School ofManagement Working Paper Series S.W.P.,December.

McGoldrick, P.J. and Andre, E. (1997), “Consumermisbehaviour: promiscuity or loyalty ingrocery shopping”, Journal of Retailing andConsumer Services, Vol. 4 No. 2, pp. 73-81.

McKenzie, S. (1995), “Distinguishing marks”,Marketing Week, 17 November, pp. 13-15.

Mazur, L. (1995), “Brands”, Marketing Business,April, p. 16.

O’Brien, L. and Jones, C. (1995) “Do rewardsreally create loyalty?”, Harvard BusinessReview, May-June, pp. 75-82.

Patterson, M., O’Malley, L. and Evans, M.J. (1997),“Database marketing: investigating privacyconcerns”, Journal of Marketing Communica-tions, Vol. 3 No. 3, pp. 151-74.

Reed, D. (1995), “Many happy returns”, MarketingWeek, 17 November, pp. 7-11.

Reichheld, F.F. (1994), “Loyalty and the renais-sance of marketing”, Marketing Management,Vol. 2 No. 4, pp. 10-21.

Reichheld, F.F. and Sasser, W.E. (1990), “Zerodefects: quality comes to service”, HarvardBusiness Review, September/October, pp. 105-11.

Rosenberg, L.J. and Czepiel, J. (1984), “A market-ing approach for customer retention”, Jour-nal of Consumer Marketing, Vol. 1, Spring, pp. 45-51.

Solomon, M. (1996), Consumer Behavior, 3rd edi-tion, Prentice-Hall, Engelwood Cliffs, NJ.

Stone, B. (1994), Successful Direct Marketing Meth-ods, NTC Business Books, Lincolnwood, IL.

Uncles, M. (1994), “Do you or your customers needa loyalty programme?”, Journal of TargetingMeasurement and Analysis for Marketing,Vol. 2 No. 4, pp. 335-50.

can loyalty schemes really build loyalty?

Documents