Matters of Taste

Sensory analysis uses taste experts to create a highly-defined model for measuring product performance. It does away with repetitive tests by predicting consumer response, says Simon Ross

Finding reliable market research methods to ensure that your products perform well is becoming ever harder, whether you are launching new brands or range extensions, or simply trying to maintain market share.

With consolidation and global expansion of corporations in many sectors, competition is fiercer, even for those manufacturers leading the way. Products that are successful in one geographical market are often quickly rolled out to others.

In food and drink, as in other industries, barriers are coming down, with successful brands in one sector launching into others. Thus confectionery turns itself into ice creams and drinks, cereals launch into countlines and spirit brands develop ever more exotic range extensions.

The growing power of stores’ own-label lines is presenting increased competition to other brands. The larger grocery chains have earned a reputation for quality and customer service on a par with many manufacturers.

The price segmentation of own label, such as Tesco’s Finest and Value lines, puts more pressure on brands, wherever they sit in a market hierarchy.

Another important change in recent years has been the fragmentation of markets, with ever larger numbers of lines on offer, as manufacturers seek to tempt consumers into the category or away from other brands.

Many of these changes are as much consumer as manufacturer-driven. Increasingly diverse work and leisure patterns mean consumers can no longer be seen as one homogeneous mass. As shopping becomes a leisure pursuit and consumers expect the pick of the world’s cuisines, the battle for a share of their rapidly diminishing attention span heats up, whether in the advertising break or the supermarket aisle.

The marketing armoury has developed in response, with ever more ingenious promotional and advertising techniques being used. However, one fundamental area which is often undervalued is product performance.

The performance of a product – whether it satisfies consumer requirements as well as or better than its competitors – will often determine sales volume. There are cases where brand leaders consistently perform poorly in blind consumer tests. However, these are exceptions and are often in highly image-led product categories.

More often, product performance plays a key role in marketing strategies. One example is in product trial. If the product is disappointing, a marketing strategy of generating trial can be counterproductive. Another example is word-of-mouth. In an increasingly marketing-literate world, word-of-mouth can be a powerful means of evaluating and communicating product performance.

Product performance can also help build up a loyal consumer base, less susceptible to competitive price and promotional activity.

Finally, consumers have become more exposed to third-party assessments of products from consumer organisations and the media, making poor product performance ever more conspicuous. Ensuring your product performs well thus becomes a necessity rather than an option to be set against other priorities.

There are a number of measures for manufacturers to be aware of. Is the repeat rate reported by consumer panel data as good as the best of one’s competitors? In image tracking, does your brand perform as well on product-related attributes as on image ones? What is the feedback from focus groups – not just from loyal users but from lapsed users or those rejecting your brand?

While these measures have value, it is prudent to approach the issue directly, by regularly assessing your product unbranded against new competitors to ensure that its consumer appeal remains strong.

Researchers trying to understand what informs consumers’ purchasing decisions are using a broader range of research methods, from accompanied shopping to video recording of product selection. Along with the more commonly-used focus groups and large-scale blind testing, these methods have to negotiate the natural limitations consumers have in articulating what motivates their choices.

Problems with consumer testing

Where price, image or even curiosity is the motivation, data can be collected from consumers with relative ease. However, the drivers of pure product preference can be more difficult to articulate, particularly in the fields of food and drink. Learning that your product is “insufficiently refreshing” or “just not the right colour” can be of limited help to even the most perceptive brand manager.

This is not simply the fault of a limited vocabulary. Some consumers may be less sensitive to the differences between products. Others may confuse sensations (acidity and bitterness, for example) or be ambiguous (is creamy a taste or a texture?). Some responses may represent the well-known halo effect, whereby consumers will score a product they like highly on a whole range of characteristics.

The classic way of overcoming this problem is to use an approach called sensory analysis. This works by separating the description of the products from the measurement of consumer preferences. Putting the two data sets together enables the relationship between product characteristics and preferences to be derived.

The basis of sensory analysis is the sensory panel. Just as some people have excellent hearing or 20/20 vision, so others have an acute sense of taste. They can taste differences where others may not. If they can measure those differences accurately and describe them in a way that is understandable to product developers, they are appropriate for recruitment onto a sensory panel.

The panel tastes a selection of products within a market, develops a range of characteristics which comprehensively describes them, and scores the products against those characteristics.

This creates a map of the market, whereby you can identify which characteristics are common to products in the market and which differentiates them. You can also see which products are similar and which stand apart.

The set of terms, objective and defined in detail, are typically adaptable to the ingredients and measurements used by food technicians while standard consumer data is not.

The greatest benefit of sensory analysis lies in overlaying consumer preference data for the products, taken typically from a hall test. By correlating the two sets of data, you can establish which of the factors are driving overall appeal. You can then see what needs to be done to improve the acceptability of each of the products.

The disaggregated nature of the data also allows other analyses. You can establish whether consumers find the same characteristics appealing or whether they are divided, with perhaps scope for niche products appealing to different segments of the target market.

Extended uses of sensory analysis

This model of any particular market, linking consumer preferences with product characteristics, removes the need for endless repeats of expensive and complex consumer testing. The characteristics of potential products can be fed into the model to predict consumer response.

Products from different batches, production lines or sites – or various points in the distribution chain – can be analysed by a sensory panel relatively inexpensively and compared with previous readings and the consumer ideal.

As marketing has become more sophisticated, developments in research have focused on brand image. Brand equity models have overtaken the more established usage and attitude studies and advertising tracking.

As with products, however, consumers find it easier to say whether they find a brand appealing than to define its image. Where an image can be defined, this is very different from establishing the sources of that image. Here, too, sensory analysis can help.

Research into hand creams has established that particular characteristics of the creams were very powerful in driving the perceived uses of the various brands and hence their overall imagery. Linking objective measures of the products with subjective perceptions of their benefits enabled the product-driven nature of the market to be clearly established. This allowed the manufacturer to position its products closely in line with consumer expectations.

Applications to brand image

A similar approach has proven effective in brand image. A highly structured method of depth interviews using extensive product stimulus is used to interview consumers from the target market and generate a range of factors which characterise the brands within the market. These factors range from typical consumers and usage occasions to more abstract images thrown up by the brands concerned.

As with sensory research, the products are scored on these attributes to give an initial impression of their image. The brands are then scored again by a larger sample of consumers from the target group, who also indicate their preferences. By correlating the preference and image scores, you can infer the drivers of brand appeal.

Although product-driven research is not as widely used as more recent observational and motivational techniques, it remains fundamental to the health of many successful brands. Sensory analysis, in particular, gets to the heart of what consumers find appealing or off-putting about a product, and places it in relation to other products in the market which share its characteristics.

Simon Ross is associate director of Mobile Sensory Testing Services (MSTS) which specialises in the product testing of food and drinks.

Research 99

Research 99 is taking place at Olympia 2, London from October 19 – 21.

It seems access to new research tools is proving the drawcard of the 27-session seminary programme. Seminar registrations have increased by at least 155 per cent compared to last year.

Research 99 is the leading market research event in Europe and addresses the needs of the industry to remain abreast of innovative and practical research techniques. Industry practitioners from companies, including MORI, Mercator and the Henley Centre, will present real-life case studies featuring well-known brands.

This year, the show will feature a Cyber Hub which will showcase the latest technologies using the Web as a research tool. The Cyber Hub will be host to research organisations which offer specific online expertise.

The Marketing Services Village will contain exhibitors offering tailored services complementary to the research process – including database management, telemarketing, direct marketing and fulfilment.