Marketing Research A to Z

Many of us probably remember the Coke Classic/New Coke marketing debacle back in 1985. A widely respected research firm had conducted blind tests proving that 60% of consumers preferred a sweeter taste that the original and that 52% chose New Coke over Pepsi. The research prompted the company launch New Coke as its flagship product. After a storm of protests, though, the original Coke formula was back as Coke Classic, and New Coke and soon thereafter Cherry Coke were relegated as sub-brands.

The fiasco taught all marketers a lesson: Research, while an essential marketing component, can lead to false conclusions if it's been erroneously or inadequately designed. Because my firm has been receiving an increasing number of inquiries about our research programs (there's nothing like an economic slump to force close examination of market positioning), I thought it would be useful to devote some space to this important topic. Remember, though, research for its own sake is worthless. You need to have a reason why you want the information and a plan for applying it to the problem at hand.

I'll start with an overview of the various research methods, culled mostly from textbooks Marketing Research by Alvin C. Burns and Ronald F. Bush and Principles of Marketing by Kotler and Armstrong. Then I'll discuss in more detail those I believe are of most use to printers.

Types of Research

In the broadest context, there are two main types of research: applied, which is conducted to solve a specific problem (e.g., What services are not meeting the needs of customers?), and basic, which is undertaken to extend knowledge (e.g., What's the market for sheetfed printing?). Marketing research is regarded as applied since it's conducted to help solve a problem or answer a question.

There are two types of collected information in applied research: Primary data, which is information being collected by the researcher for a specific purpose, and secondary data, which is information that already exists, having been collected by someone else. For today's researcher, the Internet is a gold mine for collection of secondary data. There are many commercial and non-commercial off- and on-line sources. Dun & Bradstreet, Lexis-Nexis, CompuServe, the SEC, and the Small Business Administration all have online databases. Print-specific resources include industry trade associations and online resources like,,, and While secondary data is generally easier and quicker to obtain than primary data, it may not be specifically relevant to the question at hand. In most cases, companies use a combination of secondary and primary data collection.

Many marketers overlook an obvious source of information ... internal systems. Companies with integrated accounting, sales contact management and production systems are better able to organize reports and data to glean critical information. This is another powerful argument for upgrading and integrating internal business systems. However, marketing managers can design their own internal system in the absence of Management Information Systems. Also referred to as MIS, a Marketing Information System can be developed to capture pertinent data. Of course, sales reps are another source of information about changing market conditions and competitive intelligence.

Research Methods

Primary data collection can use one of three approaches: observation, survey and experiment. The observation method is well known in the consumer product world, where people are observed in stores as they actually purchase. Surveying is the most widely used method of collecting primary data because it's the most flexible in terms of contact vehicles that can be used.

The experiment model involves taking an independent variable, such as the price your company charges for paper, and manipulating it to see how it affects dependent variables, such as the direct cost from the supplier, without influencing extraneous variables, such as transportation and storage costs. Control groups are often used in experiments to help validate findings. A control group consists of randomly selected individuals who are equivalent in age, gender, etc. to the experimental group, but who aren't subjected to the variable manipulations.

In each approach, the contact method you select will be determined in some measure by the type of research method you employ. For example, mail, phone and online methods are appropriate for surveys, whereas observation studies are limited to personal contact. The instruments you use to collect the data are also dependent on the research method. Questionnaires are ideal for surveys and observation methods, whereas experiments require more sophisticated measuring instruments.

I've written about focus groups in other columns, so I won't go into much detail here. However, in my experience they are one of the most valuable and flexible research methods available to marketers. They allow the interviewer and client to "focus" the discussion to probe topics in more depth than is possible with survey methods. There are drawbacks, however; the biggest being the fact that the sample size is small, making it difficult to draw concrete conclusions. For this reason, companies will often employ focus groups after they've conducted surveys or other quantitative research in order to dig deeper into findings.

The Internet now enables focus groups to be conducted via videoconferencing. While fast and cost-effective, videoconferencing may not be appropriate for some studies, since they automatically limit the sample population to Web users. This could be an issue, for instance, if you want information about production personnel. The technology, too, has a way to go before it matches the personal dimensionality of a live focus group where viewers watch first-hand participants' body language and interaction with each other.

Research Design

In the above Coke example, the researchers omitted critical questioning about buyers' loyalty to the Coke brand. It was a lesson that quickly elevated brand to a top priority in marketing discussion. While the term today is misused and abused, a company's brand should be viewed as synonymous with the company, making it an integral factor to be considered in all marketing research.

There are three categories of research design█exploratory, descriptive, and causal. Your research objective will determine the design. Exploratory research is used when you have very little background on the problem you need to solve or question you need to answer. There are vehicles█focus groups, experience surveys, and secondary data█most appropriate for obtaining exploratory information. Descriptive research is the most familiar, providing answers to specific who, what, where, when and how questions, like finding out how customers view our company and our competitors. Causal research uses "If x, then y" statements to gain understanding of a particular phenomenon. Used more in scientific research, it can be an effective method, for instance, to determine how buyers purchase printing.

In my next column, I'll discuss surveying methods, focusing on questionnaire development.

Charlotte Mills Seligman is president of Traversant Marketing Communications. The firm specializes in planning and executing integrated marketing programs for printing and allied graphic arts companies, with nearly two decades of expertise in the industry. Previous columns and issues of the company's Ti Monthly e-newsletter are posted on Inquiries should be directed to (415) 357-2929 or charlotte

© 2001 Charlotte Mills Seligman

April 23, 2002
Print & Graphics
Column #37, 4/02
Printing Journal
Column #37, 4/02
Research Best Practices
By Charlotte Mills Seligman

<Column Archives