New Product Management (9)

Chapter 9: Concept Testing

The importance of up-front evaluations

The biggest cause of new product failure is that the intended buyer did not see a need for the item – no purpose, no value, and not worth the price. In concept testing, we get our first confirmation that this will be a quality product. We save time by gathering information and making decisions that help assure the product will move through development fast. We lower cost in several ways (avoid the rising cumulative expenditures curve and the best time to get off a loser is at the bottom of the curve, the elimination of the many losers naturally picked up in an aggressive concept generation program, info gathering also helps make cost forecast). The highly relevant is also the stage where we set the basic marketing strategy on firm ground. We confirm the target market and settle on a product positioning statement which guides all the rest of the marketing activities.

PIC – the earliest evaluation that a firm makes is of itself and its situation. These decisions decree what types of new products fit best (ex. Nabisco sought technological breakthroughs in snack foods). The PIC will eliminate many new product ideas as firms will reject ideas that violate PIC guidelines. Following the PIC should result in excluding these ideas: ideas that require technologies the firm does not have, ideas to be sold to customers about whom the firm has no close knowledge, ideas that offer the wrong degree of innovativeness, and ideas wrong on other dimensions (not low cost, to close to certain competitor). When used at the beginning of the new products system, it precludes the unfortunate practice of having unwanted proposals eat up valuable development fund before they are detected.

Market analysis – The second evaluation that precedes appearance of the concept, it is an in-depth study of the market area that the PIC has selected for focus. This study takes place immediately after the PIC is approved.

Initial reaction – concepts begin flowing in (usually very fast), and opinions on them are formed instantly. Firms have special technique to handle this deluge more systematically (for example: if either sales VP or engineering VP approved the idea, the idea will be sent to another VP, and once both VPs approved, the idea went to a committee – both VPs primarily use their experience to judge). However, firms must resist the “bazooka effect” (where suggestions are quickly blasted out) – the idea source does not usually participate in the initial reaction (idea originator should not be able to vote), two or more persons are involved in any rejection decision, and the initial reaction is based on more than a pure initiative sense (evaluators are trained and experience and records are kept). PIC is often use in this initial reaction (knowing what a firm really wants). Most firms also use heuristics (rule of thumb) for this rough screening – evaluation are based on three factors: market worth, firm worth, and competitive insulation.

Concept testing and development – most major firms make use of concept testing (while some firms rely on intuition and experience). It is a mandatory part of the process for makers of customer packaged goods and its use is growing in industrial firms. However, there are times when this does not help – when the prime benefit is a personal sense. Concept testing usually fails in testing the aroma of new perfume or the taste of new food as the concept cannot be communicated short of actually having some product there to demonstrate (a type of gun failed badly in concept test and became very successful in real market). 2nd, concepts of new art and entertainment are tough to test (Whistler could not have concept-tested his idea for a painting of his mother). 3rd, when the concept has some new technology that users cannot visualize and 4th when firms mismanage concept testing and then blame the tool for misleading them (New Coke got favorable reply when tested and sometimes customers were asked to predict their behavior without knowing all the facts – they will deceive developers who are not careful). 5th customers sometimes simply do not know what problems they have. For service, there is also a prototype concept testing (which is much more reliable with physical prototype to talk around).

What is a new product concept?

A concept is an idea or an abstract notion – the product promise, the customer proposition, and the real reason why people should buy. It is a stated relationship between product features and customer benefits (needs) – it is a claim of proposed satisfaction. This promise can have four interpretations: producer/ consumer’s perception of the features of the new product, and the producer/ consumer’s estimate of the benefits delivered by that set of features. A complete new product concept is a statement about anticipated product features that will yield selected benefits relative to other products/ problem solutions already available.

The purpose of concept testing – concept testing is a part of the prescreening process (full screening of the idea just before beginning serious technical work. The purpose is

1) To identify the very poor concept so it can be eliminated.

2) To estimate the sales/ trial rate that the product would enjoy (market share/ revenue). The top-two-boxes score is the total number of times one of the top 2 boxes on the questionnaire were checked. Researchers usually calibrate their figures (if the total is 60%, the real figure might be 25% - while sometimes experience calibrates the probable intention higher than the respondents say now – on complex items, customers prefer to have a chance to see the final item and hear about it).

3) Purpose is to help develop the idea.

Considerations in concept testing research

Preparing the concept statement - A concept statement states a difference and how that difference benefits the customer (it sounds like a positioning statement).

1) Format – should make the new item’s difference absolutely clear, claim determinant attributes, offer a chord of familiarity (to things familiar to customers), and be short, credible, and realistic. There are different formats: a narrative format (brief, only minimum of attributes), a drawing/ diagram (usually supplemented by a narrative statement), a model/ prototype (only useful in special situations – simple to prepare food or concepts so complex that customer cannot react without more knowledge), or in virtual reality (captures the advantages of the prototype without most of it disadvantage).

2) Commercialized concept statement – presented in promotional style (a marvelous new way to chase the blahs from your diet has been discovered by General Mill scientists – a low-calorie version of ever-popular peanut butter – as tasty as ever and produced by a natural process). This format draw different reactions and produce more realistic evaluations but they also risk bias of good and poor ads writing. It should be simple, clear, and realistic.

3) Offering of competitive information – customer know much less about products and other options than we’d like. A new concept may offer a benefit that customer does not realize is new. One solution is to provide a full data sheet about each competitive product while most data will be bent slightly in favor of the new product (with full info).

4) Price – whether to put a price in the concept statement. Some object saying reaction to concept is wanted (not price). Yet price is part of a product (an attribute in customers’ eyes) and buyers cannot be expected to tell purchase intentions without knowing price.

Define the respondent groups – we would like to interview all persons who play a role in deciding whether the product will be bought and how it might be improved. Stakeholders – any person/ organization that has a stake in the proposed product. Some try to seek out small number of lead users, influencers, or large users. This saves money and gets more expert advice but often fails to reflect key differences. Some firms try to put interest in innovators and early adopters (concentrate concept testing solely on them), if this group is interested, others are likely to be so.

Select the response situation – There are two issues: the mode of reaching the respondents and if personal, whether to approach individually or in a group. Most testing takes place through personal contact (direct interviewing samples about 100-400 people or smaller), which allows the interviewer to answer questions and to probe areas there the respondent is expressing a new idea or is not clear. Groups (focus-groups) are also excellent when we want respondents to hear and react to the comments of others. The real-time response survey combines the best features of the focus groups and surveys (proven useful in screening new customer product concepts) – computers are used in real time that result can be read directly and moderator can develop original open-ended questions and ask them in real time.

Interviewing sequence – simple interviewing situations state the new product concept and ask about believability, buying intentions, and other info wanted. We first explore the respondent’s current practice (how they are trying to solve their problems, what competing products they and what do they think, how willing would they be to change and what specific benefits do they want?). This helps us understand and interpret comments about the new concept. We are especially interested in what changes they would make in the concept (what it would be used for and why, what would be replaced). We are exploring what people are doing and thinking.

Analyzing research results – most firms rely on a simple top-two-boxes score (may be adding 30% based on industry experience). Sometimes more info is needed as we cannot assume that all customers will have the same needs or look for the same benefits. Through benefit segmentation, a firm may identify unsatisfied market segments and concentrate efforts on developing concepts which suit those needs.

To identify benefit segments, customers are to rate how important each attribute was in determining their preference among brands (importance rating – used to model existing brand preference and predict likely preferences for new concepts). Cluster analysis puts observations together into relatively homogeneous groups (it is also a data reduction method like factor analysis). It groups together individuals into a small number of benefit segments.

Joint space maps – a perceptual map which allows us to assess the preferences of each benefit segment for different product concepts. The most direct way is to get customers to rate their ideal brand on each attribute. We expect the brand that is located closest to a segment’s ideal brand will be preferred by that segment. Preference regression can be used to identify the optimum combination of attributes desired by the market. This relies on a different kind of numerical input (ranking of brands). The relative sizes of the regression coefficients give us an indication of the relative importance of each factor. The ideal vector represents the optimum proportion of attributes desired by this market.

Conjoint analysis – It is useful in concept testing too (not only concept generation). Based on the three attributes of Salsa, conjoint analysis can be used to identify high-potential gaps: combinations of attributes that customers like, and are not on the market yet. The model identifies the levels of combinations from most to least preferred. Each combination could be thought of as a concept and the top-ranking concept have the highest potential and should be consider for further development. This analysis is extremely useful in concept testing because its ability to uncover relationships between attributes and customers’ preferences. Benefit segments can also be identified as the analysis identifies each customer’s value system, the relative importance of the attributes to each customer, and the preferred level of each attribute.