By Kelly Kwon, Data Analyst, Polaris Marketing Research

     There is great diversity in marketing research projects. While some studies are limited to a review of secondary data, others require complex questionnaire designs, large scale data collection, and analysis. But even with this great variability, commonalities among marketing research projects enable us to organize the process into six major steps. The value in characterizing research projects in terms of steps is that, first, it gives non-researchers and researchers an overview of the entire process, and second, it provides the researcher tasks to consider and in what order. There are the six steps of the marketing research process; here are the first three.

Step 1: Identifying and Defining Your Problem

     By the time you consider marketing research, you already have a problem that requires a decision and further exploration. Having adequate information on hand can help you hone in on the objectives by identifying the statistics needed or defining the key issues. In this step, there are two sources of information to help you define your business problem – external and internal information.

     Unless your product is a fairly new concept to your industry, chances are studies have already been conducted that relate to the information you are seeking. Obtaining external information or conducting “secondary research” means looking up information that already exists. You can search the information in the traditional way by going to the library and searching indexes, or you can take advantage of online sources such as search engines (Google, Yahoo!), trade association websites, U.S. government websites (Census, Edgar), or paid informational services (Dialog, NewsNet). Dialog.com holds 1.4 billion unique records of market research reports, trade publications, newspapers, intellectual property information, and analyst notes that provide support for decision-making. NewsNet holds many industry-specific newsletters such as Research Alert, which covers consumer trends.

     Your company database probably has information on transactional data including purchases, customer service inquires, phone logs, web clicks, and much more. Sifting through internal information, also known as “data mining,” takes time and the use of database software, but finding patterns in this data can help explain past events and also predict future events. For example, data mining can be used to identify patterns in customer’s buying behavior or identify profitable customer segments.

     Critical decisions require marketing research, but fortunately, many decisions can be made using existing external and internal information only. When considering research, you must take into account time and money, and weigh the value between information and cost. Before spending thousands of dollars and time that you don’t have on a study, you may be able to find and use information that’s already available to make your decision.

Step 2: Developing Your Approach

     Now that you have a better defined problem and an identified set of objectives, the next step of the marketing research process is to specify the approach to reach your objectives. From deciding on the best methodology to funding the project, here are questions you must answer when developing your approach:

• Project Analysis
  o How to reach the respondents and how many to reach?
  o Which methodology should be used to collect the most accurate data?
  o Will the project need advance analysis?
• Skills Analysis
  o What is my team’s marketing research skill?
  o Is the expertise available in the time frame given?
  o What can we do internally and what must we outsource?
• Budget Analysis
  o Is the information worth $20,000 or $200,000?
  o How do I stretch my research dollars? Can the information and budget be divided and used between departments?
  o When will the budget be available?

Step 3: Design Research

     Depending on the methodology, the skill set available, and the value of the information considered in step 2, designing the research requires the greatest amount of marketing research expertise, as this is the foundation of the information that will result from the project. The two main challenges in this step are the questionnaire design process and sample size.

     In the questionnaire design process, simple steps such as ordering questions can affect an accurate read of consumer insight. This is particularly important when placing the overall satisfaction question. Certain questions can trigger one small negative memory that will influence the respondent’s overall impression of the company and give a low rating for overall satisfaction. Therefore, it is best to place this question as one of the very first questions of the survey. Another important aspect of questionnaire design to consider is respondent bias. Rotating responses or using opposite wording can capture respondent’s attention and prevent respondents who select the same choice throughout the survey (straightliners). Additionally, to perform certain types of data analysis, the survey must use specific types of scale questions. For example, a driver analysis or multiple regression (used to determine “key drivers” of overall satisfaction or another single variable) is only accomplished by an interval scale, whereas a ratio scale is used when you allow zero value for categories such as age or income.

     For accurate data, you need to know the size of the entire population and the number of people you must survey (sample size) to get the data within a certain margin of error. To calculate an acceptable margin of error, first select the confidence level (how confident you can be that the error rate you find reflects errors due to chance). For example, if you have a margin of error of +/-3 at a 95% confidence level and you conduct the survey 100 times, your answers will be within +/-3 of the reported result 95 out of 100 surveys. (An easy-to-use calculator for margin of error, significance and sample size can be found in the Resource Center of our website: http://www.polarismr.com/education/tools_index.html.)

     Since the intelligence eventually gained from the research is so closely related to the selected research design, this is the single most important step in the marketing research process, and the step most vulnerable to typical research errors.

Check back next month for steps 4 through 6.

Kellly Kwon is a Data Analyst in the analytics department at Polaris Marketing Research Inc., where she handles a variety of data manipulation tasks involved in survey research. She has a bachelor's degree from the University of California, Berkeley in Molecular and Cell Biology and in Marketing from Georgia State University.