Which of the following is one of the least expensive ways to collect market research data?

• Tweet

  Inhaltsverzeichnis Show

  • The “Big Decision” Myth
  • The “Survey Myopia” Myth
  • The “Big Bucks” Myth
  • The “sophisticated researcher” myth
  • The “most research is not read” myth
  • What are the two methods that marketers use to collect primary data?
  • Which type of interview structure would be most appropriate for conducting market research with customers?
  • What is a disadvantage of using panels for marketing research?
  • Which of the following is a disadvantage of using secondary research data?

• Post

• Share

• Save

• Get PDF

• Buy Copies

• Print

Many small businesses and nonprofit organizations assiduously avoid more than a cursory flirtation with marketing research because they misunderstand what it is and what it can accomplish. Five misconceptions often dominate managers’ thinking about it:

1. The “big decision” myth. You turn to marketing research only when you have a major decision to make; otherwise it has little to do with the details of day-to-day decision making.

2. The “survey myopia” myth. With its random samples, questionnaires, computer printouts, and statistical analyses, marketing research is synonymous with field survey research.

3. The “big bucks” myth. Marketing research is so expensive that it can only be used by the wealthiest organizations, and then only for their major decisions.

4. The “sophisticated researcher” myth. Since research involves complex and advanced technology, only trained experts can and should pursue it.

5. The “most research is not read” myth. A very high proportion of marketing research is irrelevant to managers or simply confirms what they already know. Often the research is so poorly designed and written up or so esoteric that it simply ends up in the bottom drawer.

In this article I consider each of the myths and show why they are misleading. Then I suggest several approaches to low-cost research.

The “Big Decision” Myth

Too often, marketing research is deemed necessary only for decisions involving large financial stakes and in such cases should always be carried out. But research should be viewed from a cost-benefit perspective. Its costs are usually of two types—the expenses for the research itself and the amount of lost sales and lost competitive advantage caused by delaying a decision until the results are in. The benefits result from improving the quality of decisions under consideration. Any improvement, in turn, is a function of the stakes involved and how uncertain you are about the rightness of your course of action. Note that the benefits of research are proportional to the manager’s uncertainty about which way to go, not to the uncertainty about the future.

The cost-benefit ratio may often come out against research even when the stakes are high. Take the case of the restaurant manager thinking of adding a line of Mexican dishes to the menu and investing in a series of advertisements to promote this innovation. He called in a research professional to design a study of local food preferences that would show how likely acceptance of the repositioning strategy would be. Although such a study could cost several thousand dollars, the researcher determined in extended discussions with the manager that unless the survey found virtually no interest in Mexican food in the area, the manager was going ahead with the decision to add the line.

The manager was highly uncertain about the market, but he was certain that this decision was best. The researcher convinced the manager that the research expenditure was unnecessary and that the money could more productively be used to ensure that the new line got the advertising send-off it needed to have the best chances of succeeding.

On the other hand, research can be justified even when the amount at stake is not very great. This is the case whenever the research can be done inexpensively, will not take very long to complete, and will help clarify which actions to take.

These conditions often accompany advertising copy decisions. While total expenditures are small, managers do usually have two or three candidate ads, each of which seems to have potential worth. Showing the ads to a small but representative set of prospective target customers—very modest research—usually reveals one superior candidate or at least, by pointing out serious defects in one or two candidate ads, allows a narrowing of the choice. This process has a fringe benefit in that once in a while it produces extremely good suggestions for entirely different ads.

Research may also be justifiable when the stakes initially appear modest but later turn out to have been undervalued. In this regard, it is generally useful to think through the monetary consequences of making a poor decision. When one considers the possible side effects of a bad decision on such things as the organization’s reputation, its ability to attract funding and staff, and its sales of related products, the costs may be very high indeed. Such is often the case when small businesses or nonprofits venture beyond their national borders and assume that what works in their home countries will work overseas. The international community is replete with horror stories of marketing gaffes with long-term consequences that could have been avoided with a little research.

You may grant the argument thus far but then assert that there is no low-cost research to meet these challenges, that the suggestions for research made previously involve the proverbial quick-and-dirty study that may well be worse than no research at all. The only good research, you say, is a survey carefully done. This contention leads me to the second major misconception about the use of research.

The “Survey Myopia” Myth

Any reliable information that improves marketing decisions can be considered marketing research. If you take this view, many alternatives to formal survey research come to mind. Consider the entrepreneur thinking of introducing a new service but having no idea whether the target market will accept the service or, even with acceptance, how quickly it can be expected to reach break-even. If successful, the new service would yield a contribution to profits of only a few thousand dollars in the first few years. The entrepreneur could conduct a survey to reduce this uncertainty. However, to make the research 95% certain of being within two percentage points of the break-even market share figure of 10%, the entrepreneur must make the sample size 900.

An experienced survey researcher would estimate that, assuming the questionnaire and sampling plan are already designed and ignoring analysis and report preparation costs, simply completing the interviews would cost between $3,000 and $7,000. (The amount would depend on the duration and type of interviews done.)

Clearly, such research would eat up the entrepreneur’s initial years’ contribution profits. More important is the question whether the research would yield valid data in any case. That is, one should ask whether it is reasonable to expect respondents to be candid about or even to know their likely behavior with respect to this new service, especially if many feel it would not be gracious to disappoint the interviewer or the research sponsor by showing little enthusiasm for the venture.

How else, then, might the survey research objectives be achieved at lower cost? The company can try test marketing in representative markets. This approach has the virtue of not only lowering costs but yielding useful data (that is, it shows what people will do, not what they say they will do). Testing in a number of markets also allows alternative marketing strategies to be systematically evaluated.

Another low-cost approach is the commissioning of focus group interviews of 8 to 12 members of the target audience at a time. Although the results are not strictly projectable to the larger market because the groups are not randomly selected, these results do cut the cost of interviewing by a quarter or a half. Interviewers can sometimes develop richer data in the relaxed, chatty format of the focus group.

Also, the groups can at least alert management to problems with the new service that would sabotage its introduction, as they might in foreign markets. When a company uses several focus group sessions covering the range of people likely to be target market members for the new venture, officials can spot serious problems mentioned by a modest number of participants and abort a service or product launching. Elaborate probability sampling designs are simply not necessary to satisfy this objective.

The “Big Bucks” Myth

Marketing research is much more diverse than the inexperienced observer would think. There are many inexpensive alternatives to surveys:

1. Systematic observation. Managers can obtain many kinds of marketing data simply by carefully observing behavior. Retailers have found pedestrian and vehicle traffic counts to be invaluable in assessing the success of competitors’ new products or services and in evaluating new outlet locations. To gauge the effectiveness of in-store displays or packages, a staff person can record patrons’ reactions. The video cameras now commonly installed for security can also be used for this purpose.

Managers can determine the relative importance of outlet or product features by recording customers’ questions and comments in the outlets themselves. Automobile dealers and service station managers have studied their customers’ radio station preferences simply by observing the dial settings of cars brought in for service. Salespeople who regularly visit customers can record their reactions to new offers and their future purchase plans as well as take advantage of their knowlege about competitors’ plans.

Note that what distinguishes marketing research from casual observation or a “feel” for the market is careful specification of the needed observations, systematic observation, observation at random times and places (where possible, by a variety of observers), and careful recording and analysis of results.

2. Secondary sources. Industry, government, and academic reports and papers often yield data on similar ventures tried elsewhere. Reports of case studies may alert managers to fatal defects.

A medium-sized art museum, for example, learned the value of researching the experience of others when it was planning a program to expose low-income children to the art world. To launch this program, the museum director tried to kill two birds with one stone by inviting government supporters, private sponsors, and influential citizens to observe the first night’s “encounter with culture” and see firsthand what a noble enterprise they were associated with. Before proceeding, however, the director conducted an informal telephone survey of museums in other cities to determine if they had staged similar events. What she learned caused her to change her plans drastically.

Another museum that tried the very same approach reported disastrous results. It found that its supporters were horrified to observe children of low-income families hollering and tearing about in the hallowed museum corridors—quite natural behavior for children—and actually touching the statues and other works of art. Mr. and Mrs. Uppercrust, who had given $500,000 for the Jean Tinguely kinetic sculpture, were not at all pleased to see enthusiastic ghetto youths giving the Tinguely kinetics a little help. Donations dropped sharply (although no one ever said anything), and many months and careful countermarketing were needed to overcome the harm.

Secondary data may also provide information about the time needed to gain acceptance. In most cases, entrepreneurs searching through secondary sources can often make effective use of library-based computerized information retrieval systems.

3. Archival research. Valuable marketing research data are already available to many organizations and are simply waiting to be analyzed by the enterprising entrepreneur. For example, owners can glean very good insights into competitors’ advertising strategies or pricing practices from a trip to the local newspaper and from scanning past issues of the paper selected at random. Zip code data on existing charge customers can yield much about the geographic dispersion and travel patterns of a store’s or a theater’s customers and, when supplemented by census data, can indicate income, education, and other household characteristics of these customers.

4. Systematic experimentation. Many regular marketing efforts of small businesses and nonprofit organizations are amenable to experimental manipulation. For example, by varying themes in routine fund-raising mailings, a nonprofit manager can accumulate a great deal of scientifically validated information about which marketing strategies work and which do not. Also, newspaper advertisements can be varied so that the managers will learn the effects of ad size, the ratio of white space to copy, and the use of photographs.

Even if a survey must be done, managers can reduce research costs with these approaches:

• Convenience samples involving systematic querying of customer contacts, for example. Such samples are often adequate for many exploratory research purposes to identify potential marketing problems or to develop advertising or new product ideas. Indeed, many organizations miss important opportunities for collecting data from those with whom they regularly come in contact. Salespeople can easily query customers coming to an outlet, hospital officials can interview patients, and anyone can ask opinion leaders met at meetings or cocktail parties questions about current managerial concerns.

The success of such surveys depends on careful formulation of the questions in advance, recognition of biases, thoroughness in carrying out the questioning, and systematic recording and analysis of the results.

• Snowball sampling, which is the expansion of a convenience sample of present customers by a manager who asks respondents to suggest others to query. Although obviously a biased procedure, snowball sampling has several advantages. If you use the original respondents’ names in an introduction to the second set of interviewees, you can significantly improve the chances of getting their cooperation in your research. Furthermore, the demographics and lifestyle characteristics of the second sample are likely to resemble those of the original respondents. The new sample will differ in one respect only: the respondents will not be regular customers or contacts of the surveying organization.

Snowball sampling is also a unique way to find rare populations. Customers for particular low-volume products and services can often point the way to others who could be surveyed—others who already have these items and services or are thinking of obtaining them from competitors.

• Omnibus surveys, which commercial research suppliers conduct regularly. They can often include a few key questions of interest to a marketing manager. The costs per question can be kept low since several clients share them. And, if you are from a legitimate nonprofit organization, some research suppliers may, at little or no cost, include a few questions on an omnibus survey as a public service.

Sometimes volunteers and colleges are willing to help carry out such research. Many nonprofit organizations regularly have access to pools of volunteers—through local service clubs, for instance. You can assign a few such volunteers to routine survey responsibilities, see that they are trained in survey techniques, and closely supervise their work. The volunteers should be treated as professionals, not as unpaid helpers. As noted earlier, bad research is often much worse than no research at all.

Many colleges that have marketing research courses are seeking businesses to serve as field survey research cases for term projects. Again, the key is to make certain that these inexperienced researchers are carefully trained and supervised.

The “sophisticated researcher” myth

Just as marketing research need not involve complex sampling and elaborate designs, and indeed often purposely lacks randomization, so a high level of sophistication in sampling techniques, statistics, and computer analysis is not essential. Of course, executives of small businesses and nonprofits planning to undertake programs of research should acquaint themselves with at least the rudimentary principles of random sampling, questionnaire design, and graphic presentation of results.

Even when managers need high levels of sophistication—for example, if elaborate experiments or careful field study projects are being planned—they can get low-cost assistance on an ad hoc basis. Professors at local colleges are one resource. An alternative particularly appropriate to nonprofit organizations is the voluntary help of local professional researchers. Smaller companies and nonprofits contemplating extended research programs may want to ask marketing research professionals to sit on their boards of directors.

The “most research is not read” myth

Unfortunately, executives who would rather not bother with research or who subconsciously fear the results use this rationale for their inaction. Poor research certainly does occur, but when it does it is usually a testimonial to poor planning. In my experience, no piece of well-planned research has ever been rejected as unhelpful, although it may be ignored on other, often political, grounds.

How can one ensure that research will not be wasted effort? The answer rests with both the manager requesting the research and the researcher doing it. Research will be most valuable when:

1. It is undertaken after the manager has made clear to the researcher what the decision alternatives are and what it is about those decisions that necessitates additional information.

2. The relationship between the results and the decision is clearly understood. For the manager, the researcher should prepare hypothetical tables of results indicating likely outcomes for the proposed effort. Depending on the manager’s reactions to the tables, revisions can be made to bring the research on target.

3. The results are communicated well. If hypothetical results are presented, the manager should become quite familiar with the intricacies of the research design and be able to appreciate the nuances of the findings when they are finally produced.

Research need not be intimidating. It can play an important role in effective management. While research is not appropriate to all business decisions, it shouldn’t be neglected just because managers of small businesses and nonprofits entertain myths about the process. Rather, executives should be prepared to carefully analyze the particular conditions under which research of various kinds is warranted and to consider the wide range of possible low-cost designs. The payoffs from a less myopic approach to the marketing research function can be substantial.

A version of this article appeared in the July 1983 issue of Harvard Business Review.

What are the two methods that marketers use to collect primary data?

Which type of interview structure would be most appropriate for conducting market research with customers?

What is a disadvantage of using panels for marketing research?

Which of the following is a disadvantage of using secondary research data?