INTERNATIONAL JOURNAL OF RETAIL & DISTRIBUTION MANAGEMENT

Comparative Product Testing for Own-label Marketing Ogenyi Ejye Omar School of Retail Studies, The London Institute, London, UK

assessment of the perceived quality of own-label products offered by the retailers[1,6-8]. Similarly, Anon[9], concluded that own-label branding and shopping were the most significant innovation in the retail marketing area in the 1980s. Management Horizons[10] traced the history of own-label branding in the nineteenth century. Davies[11] commented that it was founded on the principle of quality at a fair price representing good value for the consumers. During the emerging marketing era of retailing, Simmons and Meredith[12] point out that many leading retailers assumed a more dominant role in the channel of distribution relative to brand name manufacturers. Several authors have indicated that this change led to own-label development activity, stimulated by changes in consumer grocery shopping behaviour (see, e.g. [2,8,13,14]). Although authoritative data reflecting the market share achieved in various product categories by own-label may sometimes be questioned, various authors[6,9,15] have shown that the own-label share of the grocery market is rapidly increasing. Murphy and Klernan[16] have quoted Taylor Nelson AGB as having reported that the leading supermarket chains now take more than 30 per cent of their revenue from ownlabels. Sainsbury’s own-label lines account for more than 54 per cent of its sales; Tesco 41 per cent; Safeway 36 per cent; and Asda 32 per cent. This improved revenue from own-labels has forced Kwik-Save to alter its retailing policy of no own-labels to introducing its first ever ownlabel brands (No Frills). There are significant variances in the own-label share of sales in different product groups and by different market observers[17,18]. Figures from Euromonitor[7] show that orange juice has more than 50 per cent own-label sales; lemonade has 43 per cent; and cola has just 13 per cent. On the other hand, Mintel[14] has own-label cola as 11.7

The UK market for groceries has always been a battlefield between the manufacturers of national brands and the retailers’ own-label products[1-3]. Food manufacturers had hoped that own-labels would be a passing fad and that, in time, consumers would rediscover their brand loyalties. But own-labels have gained an increased market penetration and are growing at a rapid rate. This experiment, conducted among the grocery shoppers in Manchester and its surrounding areas, measures consumer perception of national and own-label grocery brands in the UK. The purpose of the experiment is to access the possible differences that may exist in consumer perception of grocery quality in the UK. This purpose leads to the objectives of the study which are to: (1) go beyond the previous work by measuring more comprehensively consumer perceptions of own-label and national brands; and (2) identify consumer behaviour associated with consumer perception of national and ownlabel grocery brands. It is hoped that the findings from this experiment will provide a strategic guide to both the food manufacturers and the retailers, and assist them in effective UK food market segmentation. The findings focused on the influence of brand perceived quality; and it was tested through brand preference and choice based on product taste test. Brand preference was operationally defined as choice of the “always buy” category, and nonpreference by choice of the “never buy” category[4,5]. A review of the literature reveals a fairly substantial number of studies dealing with various aspects of own-label shopping, including the

International Journal of Retail & Distribution Management, Vol. 22 No. 2, 1994, pp. 12-17 © MCB University Press, 0959-0552

12

VOLUME 22 NUMBER 2 1994

per cent; lemonade at 42.9 per cent; and orange juice at 28.7 per cent. Market analyst, Key Note[19] attributed the rapid increase in ownlabel market penetration to: ● the actual change in brand preference on the part of grocery shoppers; ● the fragmentation of consumer demand for groceries; ● the escalating cost of advertising; and ● the institutional changes during the 1980s that have affected both the retailing and manufacturing of food products.

Grocery shoppers were identified as people who regularly buy food for the family. They were asked to rate each product on a number of factors such as fizzy, tangy, sweet, satisfying and refreshing. Respondents were also asked to say how much they would pay for each product using a scale defined from the actual price range in the marketplace. Attitudinal data were gathered on topics such as tradition, health consciousness, hedonism, involvement in the shopping process and attitude to national brand and own-labels. Sample sizes were, initially, 50 respondents in each product field using six different own-label products and at least one branded in each field, making a total of 3,150 product assessments.

Retailers’ marketing activities have also fuelled the growth of own-labels[20]. Food retailers in particular have realized that the most successful retailers will be those who manage their ownlabel programmes based on top quality products, effective packaging and strong promotion[3,21].

Limitations With only one match extracted from every two brands, coupled with the relatively small size of the mean difference between preferred and nonpreferred brands, the sample size necessary to secure significant results could be prohibitive. Given that the sample is restricted, it is better to look on the results as representing a case study that can be analysed in terms of the issues covered but which requires further verification through research involving multiple communities.

n

Retailers’ marketing activities have fuelled the growth of own-labels n Quality could be seen in terms of how retailers view their own-label programmes. According to the Private Label Manufacturing Association (PLMA) report[21], 20 per cent of the UK retailers believe their own-label products are superior to national brand equivalents, while about 62 per cent claim they are equal to national brands. Only 17 per cent think they are lower in quality. In general, better margins and image building remain the major reasons for most retailers to carry own-label products.

Testing Procedure Each participant was given a six-pack of unlabelled cups identified only by letters ascribed on them (K, L, M, N, P and Q). The manufacturer’s label and markings on the bottle caps were camouflaged, especially if the marks were engraved. In each product line, own-label and national brand were served up together. The tasting participants were asked to choose from each product pair which brand they preferred. Their ratings on specific products’ characteristics were recorded on a seven-point Likert scale (1 = poor to 7 = excellent). The dependent variables (quality and preference perception) were operated through the taste ratings.

Experimental Design The principal hypothesis subject to testing through experimentation was this: H. Own-label offers are inferior in perceived quality to the national brand. Three product fields in which there are both strong national brand and strong own-label offers (e.g. orange juice, cola, and lemonade) were selected. A random sample of grocery shoppers were asked to assess competing brands and ownlabel products in three ways: (1) blind taste (products unidentified); (2) revealed taste (products identified); and (3) revealed taste and packaging.

Data Analysis Data analysis concentrated initially on testing for significant differences between perception of individual products in the three presentations; testing for correlation between perception and choice; and for both changes in perception and acceptance with particular reference to product quality. The data set was therefore examined in

13

INTERNATIONAL JOURNAL OF RETAIL & DISTRIBUTION MANAGEMENT

detail using advanced statistical methods (factor analysis to reduce the data; and cluster analysis to group the data) (see Figure 1). The participants were grouped into: (1) those who prefer own-labels because of comparable taste; (2) those who will buy only national brands; (3) those who will buy own-label because it is value for money; and (4) those who are undecided and may buy any brand depending on price. The resulting groups were examined with regard to: product usage, brand preference, brand loyalty, attitude towards the brand and brand choice. Analysis of the data is presented in two parts. Part one deals with brand preference in the blind taste test. Part two deals with brand choice based on the introduced and/or specific perceived quality attributes in the revealed taste test. In each case, the findings relate to the aggregate classification of the shoppers’ brand preference and choice. Emphasis is placed on the differential aspects of brand attributes, and shoppers’ intended buying characteristics.

92 58

No No

National brand orange juice Own-label orange juice

80 70

No No

Basically, there appeared to be no significant differences among the national and own-label brands of lemonade and orange juice at 0.05 level. Significant differences exist between the cola brands at 0.05 level. The participants, when asked to rate the five characteristics of the brands listed in Table II as 1 = very poor to 7 = excellent, indicated a difference that was significant in 4 = “OK” (interpreted as “just enough”), voted for one characteristic on own-label cola brand. Apart from the one case, the reported differences between the national and own-label brands of lemonade and orange juice were so minor as to be not significant (Table II). Another analysis of the data, in which the “OK” response was treated as a neutral or a zero, and the “very poor” or “excellent” responses were treated as –1 and +1 respectively, in general, substantiate the percentage findings. In a blind test, the participants were not able to distinguish the taste differences between national and ownlabel brands of lemonade and orange juice, but could clearly identify and distinguish own-label from Coca-Cola.

VOJR5 VLPR1 VCOPR1 VOJPR3 VOJR4 VOCR1 VLR6

Revealed Test F'2

The revealed test with the “OK” or “just enough” category assigned a “4” value was scored mainly for own-labels in the three product categories. Most participants tended to rate all the national brands as mainly very good. When the price and packaging were also revealed, own-label brand of orange juice was rated as “very good” (Table III). The revealed test clearly indicated that shoppers assigned superior ratings to national brands. Own-label offers were rated much lower in the revealed taste test than in the blind test.

VCOR4

VOJPR2

National brand lemonade Own-label lemonade

Table I. Blind Overall Taste Test (All Participants)

F'1

VOJUCE

Yes Yes

National brand cola Own-label cola

As a group, grocery shoppers could not distinguish the taste differences among the brands of lemonade and of orange juice. They were able to distinguish the taste of Coca-Cola from ownlabel cola (Table I).

V1FMND

116 33

Soft drink brands

Blind Test

VCOLA

Overall ratings

Significantly different from other brands (0.05 level)

VLR4

VOPR3

VLR8 VLR3

Figure 1. Rotated Factor Solution (Kaiser Normalization)

14

VOLUME 22 NUMBER 2 1994

Characteristics

VK

VL

VM

VN

VP

VQ

*

Sweetness Fizziness Tangy Refreshing Satisfying

69 68 10 42 66

16 30 07 13 17

50 40 31 34 43

34 39 14 24 33

20 10 60 11 51

18 05 45 35 44

VL (only) No VL (only) VL and VP VL (only)

* Significant differences among brands at the 0.05 level. Note: The three-place neutral centre scale is probably in need of further testing and comparison with seven-position scale to help determine the amount of bias it may have introduced.

Table II. Blind Taste Test – Specific Characteristics (All Participants)

Brand

Revealed test

Significant difference between blind and reveal test

116 33 92 58 80 70

121 26 98 76 87 56

No Yes No No No No

No

Yes

Blind test

VK VL VM VN VP VQ Significant differences between brands

and own-label brands. The results of the tests clearly reject the hypothesis of no perceptual differences. Individual scale differences found in the taste experiment provide a description of the perceived differences. When own-label is identified and/or seen it is perceived to be inferior to the national brand. However, the basic finding of the blind test indicated no significant differences between the national and own-label brands of lemonade and orange juice. Apart from price and packaging differentials, one could easily be substituted for the other. Almost 67 per cent of the shoppers who indicated via their ratings that they preferred ownlabel in the blind test changed their minds in a revealed test. Of the 67 per cent in the blind test, only 15 per cent preferred own-label brands in a revealed test. This suggests that shoppers in a revealed test did not prefer the national brands because of taste or flavour. Product choice was made on factors other than taste. When the price element was introduced, the intending choice of brands became closer. The change of minds in the revealed test was due to brand image rather than the taste. In two out of three product lines tested, ownlabel brands rated equally to national brands. In lemonade and orange juice, own-labels were regarded as value for money, and price alternative to national brands. The test, however, revealed some slight quality variations in favour of national brands. Coca-Cola was perceived as superior in characteristics measured as compared with cola. These findings are similar to the taste-on-test experiment conducted by Which magazine[15]:

Note: All brands were not significantly different from all others at the 0.05 level except VK and VL relative to each other.

Table III. Comparative Taste Test – Blind vs. Revealed (Overall Ratings)

Interpretation of the Findings The experiment produced a number of useful findings. More specifically, evidence was available to answer the following questions: Could grocery shoppers in general distinguish own-label in a blind test? ● What influence would brand identification have on shoppers’ evaluations of own-label offers? ● How far do grocery shoppers relate price to product’s perceived quality? ● Do grocery shoppers see own-labels as inferior to national brands? The respondents in this experiment have reported clear differences in their perceptions of national ●

…people think of own-label as cheaper but the result shows that people like them as much as the famous names.

15

INTERNATIONAL JOURNAL OF RETAIL & DISTRIBUTION MANAGEMENT

Retailers must constantly strive to match customer needs with brand offerings, in accordance with retail marketing policy. A shopper requiring high levels of food quality, image, prestige and confidence may be best served by a retailer offering a full range of national brands. Similarly, those who want value for money may be best served by offering a full range of own-labels. In general, better marketing and good managerial sense will ensure success.

The Which result was most significant, because Heinz, whose baked beans performed poorly in the test, do not supply own-label products to retailers.

Summary and Conclusions The results of this experiment show how ownlabels are entering a new marketing phase in their evolution. The indication is that opportunity for profit and store differentiation will continue to drive the success of own-labels in the future. The most successful retailers will, therefore, be those who manage their own-label programmes based on good quality and effective packaging.

n References 1. Senker, J., “Technological Co-operation between Manufacturers and Retailers to Meet Market Demand”, Food Marketing, Vol. 2 No. 3, 1987, p. 38. 2. Shipley, D., Hooley, G. and Wallace, S., “The Brand Name Development Process”, International Journal of Advertising, Vol. 7 No. 3, 1988, pp. 253-66. 3. Uncles, M.D. and Ellis, K., “The Buying of Own-labels”, European Journal of Marketing, Vol. 23 No. 3, 1989, pp. 57-70. 4. Liversey, F. and Lennon, P., “Factors Affecting Consumers’ Choice between Manufacturer Brands and Retailer Own-labels”, European Journal of Marketing, Vol. 12 No. 2, 1978, pp. 158-70. 5. Allison, R.I. and Uhl, K.P., “Influence of Beer Brand Identification on Taste Perception”, Journal of Marketing Research, Vol. 1 No. 3, 1964, pp. 36-9. 6. Ellis, K., “An Investigation of Private Label Purchase Behaviour in the Packaged Grocery Market”, PhD thesis, London Business School, 1988. 7. Euromonitor, “Hypermarket and Superstores Retailing in the UK”, Retail and Distribution Surveys, Euromonitor, 1992. 8. Omar, O.E., “Grocery Shopper Behaviour and Retailers’ Own-label Products”, PhD thesis, Manchester Metropolitan University, 1992. 9. Anon, J., “Value of Own-label at Sainsbury”, The Grocer, 6 July 1991, p. 18. 10. “Private Branding in Marketing Era”, Management Horizons, The RIS Retail Focus Series, 1985. 11. Davies, G., “The Two Ways in which Retailers Can Be Brands”, International Journal of

n

Retailers may want to know how consumers perceive own-labels before making decisions n Own-labels were psychologically positioned below the national brands on many of the scales, but were reported to be particularly good value for money. The experiment has identified two distinct market segments in grocery retailing, based on purchase behaviour and product choice characteristics. These findings may help to accomplish the development of appropriate marketing strategy at both the manufacturing and retail level. While only a few of the shopperbehaviour dimensions tested here distinguish national and own-label high-raters, significant perceptual differences exist in consumers’ minds concerning the two types of brands. Apparently, both the manufacturer and the retailer have achieved their objective of securing unique market positions. This experiment helps to define the unique perceptual positions of the brand categories. This information can be used as an aid in making a variety of retail marketing decisions. Retailers may want to know how consumers perceive own-labels before making decisions concerning the product mix offered for sale. They can also apply the findings of this experiment to decisions involving whether to stock own-label or not. Since own-label brands are increasing their penetration of the grocery market, the shopping process associated with their purchase may be reflecting a change in food consumption.

16

VOLUME 22 NUMBER 2 1994

12.

13.

14. 15.

16.

Retail & Distribution Management, Vol. 20 No. 2, 1992, pp. 24-34. Simmons, M. and Meredith, B., “Own-label Profile and Purpose”, paper presented at Institute of Grocery Distribution Conference, Radlett, IGD, 1983. de Chernatony, L., “Consumers’ Perceptions of the Competitive Tiers in Six Grocery Markets”, PhD Thesis, City University Business School, London, 1987. Mintel, “Own-label in Packaged Grocery Retailing”, Mintel, 1991. Consumer Association, “Taste-on-Test”, Which Reports, Consumer Association, November 1992. Murphy, C. and Klernan, P., “Countering the Own-label Brigade”, Marketing Week, 14 May 1993.

17. Davies, G. and Bell, J., “The Grocery Shopper: Is He Different?”, International Journal of Retail & Distribution Management, Vol. 19 No. 1, 1991, p. 25. 18. de Chernatony, L. and McWilliams, G., “The Varying Nature of Brands as Assets”, International Journal of Advertising, Vol. 8, 1989, p. 339. 19. Key Note, “Retailing in the UK”, Key Note Publications Ltd, 1992. 20. Rigaux-Bricmout, B., “Influence of Brand Name and Packaging on Perceived Quality”, in Mitchell, A. (Ed.), Advances in Consumer Research, Vol. 9, Association for Consumer Research, Chicago, IL, 1981. 21. PLMA, “The Future of Private Labels”, The Grocer, 2 October 1993, p. 10.

17