Frequently Asked Questions
With more than 25 years of constructing, executing, analyzing and presenting countless B2B market research projects in multiple markets, sometimes we feel we have heard it all. As a result, we decided to share some of the more frequently asked questions we often hear during the course of our projects. We are sure we have yet to hear all of the questions surrounding market research so we will continue to add more as we encounter them. Enjoy.
Although B2B market research is not inexpensive, acting upon assumptions or bad information can be much more costly. And, on the positive side, not conducting research may miss business opportunities at specific customers or market segments.
As an example, PMG completed a study for a client that had been extremely successful in producing products for the automotive industry. They felt they could leverage their assets and experience to sell the same type of product at retail. PMG investigated the opportunity and found that while the market was large, the competition was intense and channel power was held by the retailers. Consequently, margins were extremely low – much lower than with OEM customers. The client decided not to enter the market and the CEO remarked that the money spent on the research was “some of the best money I ever spent.”
Business-to-business (B2B) and business-to-consumer (B2C) markets differ in fundamental ways. Although both are comprised of producers moving products and services through distribution to sellers, the manner in which transactions happen is distinct. Major differences are summarized in the table below.
There are several key differences between consumer and business-to-business markets that impact how research should be conducted. Among them are:
- Sample size determination,
- Data collection methodologies,
- Questionnaire design, and
- Use of incentives.
Qualitative research generates textual data (e.g. comments) while quantitative research provides numerical data (e.g. scores or ratings). Qualitative research is exploratory and is often used to gain insight into the perceptions and motivations behind observed actions.
In contrast, quantitative research generates numerical data such as ratings, average scores, response frequencies. It defines attitudes and opinions using statistical inference, generalizing results from a sample to the larger population.
Some companies have come to believe that customer surveys are passé, or that the business knows what customers think because sales and customer service personnel talk with them every day. PMG has been conducting customer surveys for over 25 years, and consistently finds that a good customer survey provides insight into: business performance – compared to customer expectations and to competition, improvement priorities, and loyalty.
A focus group is a small group discussion moderated by a trained facilitator. It is designed to understand opinions on a selected topic, explore research hypotheses, identify correct terminology about the topic, explore potential segments, and often, to frame future research.
An in-depth interview (frequently abbreviated as IDI) is a qualitative, somewhat intensive individual conversation exploring opinions and perspectives on an idea, product, market or other topics.
CATI stands for Computer Assisted Telephone Interviewing. CATI is a commonly used method of data collection for large-scale studies in the B2C and B2B environments.
There is no typical response rate. The response rate will vary based on the complexity of the interview, method of data collection, subject of the survey and many other factors. Response rates can range from 2-3% up to 80%+.
The sample size is determined using appropriate formulas. There are three variables that drive the sample size:
- Size of the population,
- Range of responses/measures, and
- The desired level of confidence.
If the sample is a random sample, then the FAQ about determining sample size will address this question. However, in many B2B situations, the total population may be quite small so there is not a need to infer something about a population, and no consequent need to draw a sample.
Attributes can describe a product, service, idea/concept or company. Product and service attributes are typically drawn from marketing and promotional material. Attributes describing a concept can be more difficult to develop and may be done through focus groups, company marketing material, social media, ideation with customer-facing employees, and other methods.
If a list is being developed for a Customer Insight® survey, the client must develop the contact list. This may be downloaded from a CRM or compiled from lists held and managed by sales representatives. For market studies, the researcher can purchase a list from a list broker that closely coincides with the target market. These lists are pulled from website or magazine subscriptions, product warranty registrations, membership rosters, lists of seminars or conference attendees, and many other sources.
Scale construction is based on years of experience and empirical data confirming the use of particular scales in certain circumstances. Scales can be tested for reliability using proven statistical methods. It is best to leave scale design to market research professionals with appropriate experience.
Statistical significance can be thought of like shooting at a target with a rifle. There are many variables that influence where a bullet will hit a target but he best shooters will place their shots within a tighter circle than others. Similarly, a survey measure – if repeated – will result in a slightly different result each time. However, all the results will be within a range (or circle) of the true population value. Statistical significance is an indicator of the confidence one can place on a measurement. It is comprised of two elements: the boundary around the measurement and the likelihood of an error.
see also the FAQ “How do you determine sample size?”)
Conjoint analysis is a survey technique that forces respondents to make choices between features or products just as they would in an actual purchase decision. As he/she does so, the respondent provides implicit information about the value they place on each feature (e.g. color) and the level of each feature (e.g. blue, green, yellow). Conjoint can be used to estimate preference and, by combining results across respondents, estimate market share.
The Van Westendorp pricing model is based on a series of four questions designed to establish boundaries around what would be considered a fair price for a product or service. By looking at the percentage of respondents that believe a given price is too high or too low, the model suggests that the best price point can be established.
Ethnography is the systematic, observational study of people and cultures. The ethnographic researcher seeks to understand how people live their lives by observing people in their natural settings – in the home, as they shop, or as they use a product or service.
Innovation is a substantial challenge. One study showed a 17% success rate of innovation. The Vision Critical blog site reports that about 85 percent of new products fail in the consumer market. They also cite a study from Harvard Business School that estimates 95 percent of consumer products fail. In the face of those long odds, accurate research is vital.
While many surveys report only on performance, PMG methodologies are built around four key aspects of the customer relationship: performance, importance, competitive position, and loyalty. These four pillars define the customer-vendor relationship and can be used to isolate specific areas for improvement.
Competitive performance can be evaluated in several ways. PMG generally prefers to use the concept of the best alternate supplier in which the respondent chooses their best option to the study sponsor and rates their performance head-to-head against the sponsor. Two-thirds of the time, respondents will identify their choice of best alternate supplier. This allows a direct comparison of each attribute.
The net promoter score is a measure of loyalty. It is the result of one question, “How likely are you to recommend [COMPANY NAME or BRAND] to a friend or business colleague?” The rating is captured on a 10-point scale where a rating of 1 indicates not at all likely, and a rating of 10 indicates highly likely. The NPS is calculated by first determining the percentage of respondents choosing each point on the scale. Then, the total percentage of those rating a 1 through 6 (termed “detractors”) is subtracted from those rating a 9 or 10 (termed “promoters”). The subtraction is the “net” element of the score, promoters, net of detractors.
So, the NPS could range from -100 to +100. A score of 0 would indicate that the percentage of promoters is exactly equal to the percentage of detractors. As it turns out, the average NPS reported by most companies conducting such studies is about 20. In general, a positive score (above 0) is considered good (since promoters outnumber detractors), 50 is excellent, and above 70 is considered world-class.
To improve an NPS (Net Promoter Score), we must first understand what drives the score. The only way to do that is to find variables that are closely linked to the recommendation question that is the basis for the computation of the NPS. Improving performance on those few attributes will have the largest impact on NPS. PMG has found that the top 4-5 attributes account for over 90% of NPS improvement potential.
There are four steps to determining a market size:
- Define the market
- Conduct desk research
- Conduct primary research
- Triangulate results
Customer mapping is the exercise of identifying all buying locations within a defined geography and determining their potential volume.
A buyer persona is a fictional representation of a market segment. Personas help to visualize customers as real humans. Personas should be based on solid market research and insights from internal customer data. A typical business will have between two and ten personas.
Segmentation is the process of finding groups of customers that behave and think similarly. The criteria for establishing segments are based on marketing and sales needs, attributes that describe or limit product usage, and – importantly – the type and scope of data that are available.
Segmentation is a term applied to clusters of similar individuals or businesses comprising a market – including customers and non-customers. Classification is a term applied to customers only. Segments are used to identify potential customers that would benefit from a product or service and could be served profitably. Classifications are used to understand different groups of customers and how they can be better served by sale, technical and customer support.
The importance is an indicator of the relative visibility of each attribute in the purchase decision. It can be evaluated by asking directly (stated importance) or by computation (derived importance).
Competitive intelligence (CI) is a very large umbrella that covers many types of analysis techniques. Common forms of CI include market share analysis, financial analysis, sales process, go-to-market strategy, pricing strategy, and others.