CONCEPT TESTING (PRODUCT TESTING)
Concept testing or Product Testing is very important research technique used in Marketing Research it provides us estimate and evaluate a concept’s market potential and supplies information important to make the concept robust and developing introductory marketing strategy. Testing of product will carry out by exposing the product Idea to consumers and measuring their response and reaction towards the product, using a predetermined series of questions designed to measure various reactions, feelings and opinions.
Concept testing has several important purposes.
First, a concept test should enhance the efficiency of the development process because only concepts judged to have sufficient competitive strength and market potential pass this particular gate and go on to the next stage, thus conserving product development resources.
Second, the fully elaborated concept that emerges from the test provides direction in terms of a specific protocol for the development team to follow through subsequent development stages.
Third, the concept test helps the development team gain understanding of consumer reaction to the concept and its components.
Fourth, concept testing allows an estimate of purchase intent, and thus provides data for early stage sales volume estimation.
Fifth, concept test results provide diagnostic data as well as guidance for positioning strategy.
Concept testing can use either Qualitative or Quantitative techniques. Qualitative techniques, such as in-depth interviews and focus groups, can provide a great deal of useful knowledge regarding consumer reaction to the concept, the concept’s strengths and weaknesses, and recommendations for improvement. Qualitative research, however, cannot provide data necessary to make early sales volume estimates.
The most realistic purpose of a testing is to present the product Idea realistically to check the acceptability and willingness to buy the product. The testing yield numerous important information. Not all of this information will be yielded by any one test, and what information the test provides will be dependent on the research methodology, specifically the types of questions asked, as the proper execution of the test.
Virtually every quantitative concept test will contain an intention to buy or purchase interest question. Most will attempt to measure uniqueness and price value, and most will attempt to elicit consumer reaction not only to the concept as a whole but to the elements or characteristics that comprise the new product.
A properly planned, executed, and analyzed concept test can provide several important pieces of information:
An early read on sales potential. If potential is adequate, test results should aid in identifying the segments of the market that are potentially most responsive to the new product. If results are below expectations, the test results should permit determination of any potential targets of opportunity if the market population were to be segmented into different or narrower demographic, behavioral, or attitudinal groups. Alternately, we may find simply that the concept lacks sufficient appeal to be a viable entry into the category.
The degree to which the message that consumers receive from the concept (in both objective and subjective dimensions) is consistent with the message intended. This is diagnostic element that lets the development team know if the concept’s level of success or failure is attributable to the attractiveness and strength of the new product idea, or is compromised in some way by a biased or inaccurate execution. This is especially important in the case of a poor-performing concept when a decision must be made to either kill the idea or make another attempt at positioning and describing the concept.
Identification of the individual strengths and weaknesses of the concept, providing insight into the relationship between consumers’ overall evaluation of the concept and the role of each of the concept elements in that evaluation. The test should relate which elements of the concept contributed to positive evaluation and which contributed nothing or to negative evaluation. This insight aids in strengthening the concept via heavier emphasis on positive characteristics and elimination or downplaying of neutral and negative elements.
The types of data provided by a concept test.
Consumer attitudes and usage patterns in the product category
The competitive setting or market structure in the category
Size of the potential market
Segments and characteristics of the consumers making up each segment
The main idea being communicated by the concept
The importance of the main idea to consumers
The concept‘s relevance to consumer needs
Consumer reaction to product attributes and features, and the relationship between consumer reaction to product attributes and overall concept rating
Purchase interest and intent
Reasons consumers give for purchase interest
Major strengths and weaknesses of the concept
Strength of the concept versus other concepts tested
Perceived advantages and disadvantages relative to competitive products
Uniqueness—a measure of competitive insulation and a strong indicator of concept strength it combined with high purchase interest
Believability, the extent to which consumers feel that the concept can deliver the benefits promised
Perceived price-value—consumers expect good value so they are not necessarily turned on by concepts which provide a good value, but a poor price-value relationship can dramatically depress overall appeal
Usage situations and frequency of usage
The expected frequency of purchase
Potential source of volume (incremental category sales, from competitors, cannibalization of the developing firm‘s product line)
METHODOLOGY
The methodological issues for concept testing are essentially the same as for any marketing research project, and include type of sample, sample size, method of administration, questionnaire design, and analytical techniques.
TYPE OF SAMPLE
The type of sample can be either probability (random) or non-probability (judgment, quota, or convenience). Random samples use alternative sampling methodologies, including simple or one-stage random samples and multi-stage sampling designs, including stratified random samples, area samples, and cluster sampling. The major advantage of random samples is the ability to use techniques of statistical inference, and to determine the precision of our sample estimate or the amount of sampling error. Random samples require sampling frames (lists of potential sample members) and the use of specific and correct techniques of sample selection to guarantee true randomness. The disadvantage of random sampling is high cost relative to other sampling designs.
Non-probability (or purposive) samples are of three major types. The first type is the quota sample, whereby we use quotas to draw a sample that matches, on certain predetermined dimensions, those same characteristics of the population. The second type of nonprobability sample is called a convenience sample. A very typical convenience sample design is administered using mall intercepts (central location testing), where the focus is on conveniently (and efficiently) reaching a large number of potential respondents. The third type is the judgment sample, which involves selecting sample members who, because of their background and/or experience, should be a useful source of information. The purpose of a “purposive” sample is to draw a sample which is representative of the population of interest, which in concept testing is the potential market for the new product.
Companies conducting their own tests typically use either central location tests using a combination of convenience and quota sampling, or use mail surveys sent to households or individuals who are members of a preselected and demographically balanced panel. However, since concept testing, like the entire process, is iterative, one project may feature multiple concept tests using different types of samples with concepts that become more elaborate and more focused with repeated testing.
SAMPLE SIZE
There are no hard and fast rules governing sample size. The old rule in basic statistics is that if we are using a probability sample, we can use a z-test instead of a t-test if the sample size is 120 or above. The tradeoff in sample size is, of course, cost versus representativeness and accuracy. A more representative sample requires a larger sample size as the population becomes more heterogeneous. Most concept tests use samples consisting of 300 to 500 potential respondents. This number is necessary to provide adequate size of each cell when doing cross-tabulations. Smaller sample sizes can be used for relatively homogeneous populations. For instance, the Tasty Baking Company launched Tropical Delights, which is targeted to Hispanics and specifically Hispanics of Puerto Rican extraction. The relative homogeneity of that population permits a smaller sample size to yield useful and projectable data.
ADMINISTERING THE TEST
Concept tests can be administered in person, typically via the use of central location tests or focus groups; by the use of mail questionnaires; and by the use of telephone surveys, Online Surveys. In-person interviews allow a great deal of range and flexibility in the types of questions researchers can ask and the visual stimuli that can be employed, and typically allow for greater amounts and depth of data to be gathered. In-person interviews are expensive and require trained interviewers as well as a location in which the test is administered. Central location tests require careful screening of potential respondents to ensure sample representativeness. In a depth interview situation, researchers strongly prefer personal interviews. However, concept tests in many cases use closed-end questions and scaling questions, which can be effectively and efficiently handled via telephone, Online or by the use of mail surveys.
Online interviews allow a great deal of control, access to a greater range of respondents in a shorter period of time than central location tests, Online surveys tend to be the least expensive, provide an opportunity for the respondent to ponder questions and even gather information before answering questions, allow for visual stimuli to be presented, and allow for careful targeting of households or at least neighborhoods.
DESIGNING THE QUESTIONNAIRE
A typical concept test questionnaire would begin with an assessment of the respondent’s current experience and practice in the product category, demonstrate the concept, elicit reaction to the concept, measure intent to purchase, and then ask for relevant demographic and other data useful in classifying respondents. Table 6.2 lists several dimensions on which reaction to the concept can be assessed.
The most important question in concept test is the purchase intent question; this is the tool most frequently used to measure concept success, both as a decision-making variable, and as a key element in various volume-prediction models. By asking this question, we not only measure the overall strength of the concept, but we also can pinpoint the type(s) of individuals most likely to buy the product. A typical statement of the purchase intent question is:
If (this product) were currently available in your local store, how likely is it that you would buy/try this product?
Since most available normative data are based on a five-point scale, the purchase intent question usually employs such a scale with the following scale points:
• would definitely buy
• would probably buy
• might or might not buy
• would probably not buy
• would definitely not buy
The percentage of respondents who respond “definitely would buy” is called the top box score. The percentage of respondents who respond “definitely would buy” plus the percentage that respond “probably would buy” is a measure of positive purchase intent and is called the top two box score.
• For diagnostic purposes, that is to learn whether the basic idea of the product has merit or needs improvement, the purchase intent question should be asked twice, once at the beginning of the survey with no price and again toward the end with price. This is based on the premise that the respondent can provide better diagnostic information without reference to price.
• Because the reasons consumers give for having high or low levels of purchase interest often provide valuable insights about a concept’s strengths and weaknesses, we should assess reasons for purchase interest. This can aid in spotting winners and losers, and in understanding the factors responsible for their performance.
• Analyze the reason for purchase interest separately among individuals with positive or negative purchase interest. Also, respondents with neutral purchase interest often like the product on some dimensions and dislike it on others, and the elements they dislike may be different from the things disliked by those with negative purchase interest. Therefore neutral respondents should be analyzed separately provided that the number of people who fall into this neutral group is large enough to be significant analytically.
Purchase Intent alone is not sufficient to properly evaluate the concept.
Other key indicators that can make a big difference in the value of your test data.
The concept’s uniqueness
The concept’s appropriateness or relevance to the consumer’s needs
The main idea being communicated by the concept
The importance of the main idea
The expected frequency of purchase
Uniqueness represents an important dimension for an obvious reason: successful new products must be meaningfully different than competitive products.
Uniqueness is an excellent diagnostic tool to help pinpoint how to improve a deficient concept. Cross-tabulating uniqueness by purchase interest can help pinpoint factors that are responsible for weak concept ratings.
Determining whether and to what extent consumers perceive a concept as delivering important benefits or satisfying recognized needs is particularly important in understanding whether consumers might adopt a new product.
Often, we are willing to buy a new product once or twice, because novelty is an important motivation, but also because new products often come with significant incentives that lower the effective price. But one or two purchases do not sustain a new product. Unless we are specifically planning to launch a “fad” item, such as a ready-to-eat breakfast cereal using a licensed character, or a seasonal item, such as a Christmas snack cake, we need consumers to make the product part of their regular lifestyle and consumption system. This can only happen if consumers perceive the item as having the ability to meet ongoing and important needs.
Virtually every concept test questionnaire includes a question that asks respondents what they consider to be the main idea of the product. Purchase intent and attribute ratings can only be valid if consumers understand the concept.
Other useful findings from concept tests include identifying the user and the usage occasion(s), and how the concept might be improved. The user is the type of person that respondents associate with your product. This information is important primarily in planning the advertising because user imagery is a key element of creative strategy, particularly in categories such as beverages, cosmetics and fragrances, and tobacco products. Usage occasions not only help to identify the competitive set, which may and often does differ according to usage occasion, but also provide clues for our positioning and marketing program.
Consumers often can provide very good insight into how to improve a concept.
Simply include the question, “In what ways, if any, could this product be changed or improved?
ABSOLUTE VERSUS RELATIVE CONCEPT TESTS
Concepts can be tested in two ways; either individually or with control concepts. Most experts prefer absolute or monadic tests as opposed to relative or sequential tests. This is because absolute measures are less biased and more amenable to sales volume estimation.
VOLUME ESTIMATION
One of the most critical outcomes of the concept testing phase is an estimation of first-year sales volume.
Market size estimates are based on category data that measure the number of households purchasing items in the category plus a projection of the geographic scope of the launch. For instance, canned soup has close to 100% household
Data Requirements for Estimating Sales Volume for a New Product.
Estimated market size, usually stated in terms of number of households in the target market
Projected level of awareness of the new item which will be created primarily through consumer advertising plus consumer and trade promotions
Projected level of distribution penetration, typically stated in terms of a percentage of ACV distribution to be attained within a given period of time after launch, and based on both the strength of the concept and the strength of the trade program
Estimated percentage of households that will try the product plus the average trial volume per household
Estimated percentage of households that will buy the product at least once after trial (repeat purchase) plus the projected number of purchase cycles and the average volume per purchase during the introductory period
The price of the product to the trade
Awareness levels are projections based upon relationships between the strength of the introductory marketing campaign and the percentage of households in the target market that will become aware of the new item. Introductory marketing campaigns, for purposes of these projections, typically consist of several different elements. Media advertising campaigns help build awareness. The strength of an advertising campaign or program is usually measured by the number of gross impressions delivered (i.e., the number of exposures of the message to consumers, including multiple exposures to the same consumer) or by gross rating points. Gross rating points (GRPs) represent a measure of the weight of an advertising campaign. GRPs are calculated by multiplying the net unduplicated reach of the campaign, stated in terms of a percentage, by the frequency, or average number of times each household is exposed to at least one message.
CAUTIONS
• The concept statement is a description written to explore potential consumer interest; include some “sell,” i.e., how the product is different, makes life easier, etc.
• Include some sort of price context to avoid major misunderstanding.
• Don’t make materials too elaborate; retain flexibility in presentation.
• Be prepared to modify the benefits as you get consumer reaction.
• Consider “layering” different options on the basic concept.
• Elicit individual as well as group opinion.
• Make sure you understand the difference between lack of consumer understanding (i.e., they understand the concept differently than us) and consumer confusion (they don’t know what we are talking about)
SUMMARY
Product concepts are verbal or pictorial representations of products. Concept statements provide a description of a new product’s attributes and benefits, as well as a stimulus to which consumers can respond to assess the overall strength of the concept and the level of consumer interest. Concept statements also provide a protocol to guide the work of the project team as the concept is developed into a physical product. Concept tests employ a variety of research methodologies to measure consumer reaction to the concept, determine the concept’s strengths and weaknesses, and measure purchase intent, and serve as a gate through which only high-scoring concepts pass to the next stage of development.
BIBLIOGRAPHY
JOHN B. LORD.2000 “Product Concepts and Concept Testing”, by CRC Press, Inc..
Clancy, K. and R. Shulman. 1995. “Test for Success.” Sales and Marketing Management, pp. 111-
114.
Clancy, K. J., R. S. Shulman, and M. M. Wolf. 1994. Simulated Test Marketing. New York: Lexington
Books.
Crawford, C. M. 1997. New Products Management. Boston, MA: Irwin McGraw-Hill.
Feig, B. 1993. The New Products Workshop. New York: McGraw-Hill.
Fitzpatrick, L. 1996. “Qualitative Concept Testing Tells Us What We Don’t Know,” Marketing News,
30(20):11.
Fuller, G. W. 1994. New Food Product Development: From Concept to Marketplace. Boca Raton, FL:
CRC Press.
Gibson Consulting Co. brochure.
Gill, B., B. Nelson, and S. Spring. 1996. “Seven Steps to Strategic New Product Development,” in
The PDMA Handbook of New Product Development, Rosenau, M. D., A. Griffin, G. A. Castellion,
and N. F. Anschuetz, eds. New York: John Wiley & Sons, pp. 19-33.
HBS Case Services case number 595-035, Nestle Refrigerated Foods: Contadina Pasta & Pizza (A)
Rev. 1/97.
Patrick, J. 1997. How to Develop Successful New Products. Lincolnwood, IL: NTC Business Books.
Schwartz, D. 1985. Concept Testing, New York: AMACOM.
Stinson, W. S., Jr. 1996. “Consumer Packaged Goods (Branded Food Goods),” in The PDMA Handbook
of New Product Development, Rosenau, M. D., A. Griffin, G. A. Castellion, and N. F.
Anschuetz, eds. New York: John Wiley & Sons, pp. 297-312.{~?~page>129{~?~/page>