How To Create a Winning Questionnaire


There is a wide array of survey software, with a range of features and options that make it easy for businesses and organizations to get feedback from their target audiences. Some survey software tools are very simple to use while others are more complex and sophisticated. Most have free versions that provide very basic survey tools, are available online and hosted through secure cloud servers. Some of the most popular include Survey Monkey, Survey Gizmo, Typeform, Google Forms, and Zoho Survey.  Selecting the one tool that is right for you, from the hundreds that are available, will depend on your needs and your budget.

Even though software has made it easier to collect and analyze customer data, the quality of that data is still dependent on the questions asked. Good questions yield good data while bad questions result in bad data.

More often than not business owners will begin a research effort by formulating questions. And of course, they don’t want too many questions because nobody likes long surveys. What usually results is a list of 3 to 5 Yes/No questions that are more than likely related to “nice to know” rather than “need to know” aspects of their business. With this sort of approach it’s no wonder that they are often disappointed with the final results and end up being skeptical of the value of surveys.

This is why survey Do It Yourselfers must begin by first establishing objectives for their research and then translating the objectives into questions that can be asked in a survey. If you are to the point where you have decided to conduct a survey then obviously you want to get the most accurate unbiased information possible to support your decision-making.

Survey Objectives

Survey objectives should be used to determine the type and number of question(s) that will be asked. The objectives will also help in identifying who the survey respondents will be. Each question you come up with should be evaluated against the objectives. Using  survey objectives as a filter will also help in prioritizing and selecting the questions that will be on the final survey instrument. If a question is not related to one of your research objectives it should not be on the survey. This step will increase the likelihood that your data will be actionable and meaningful.

Question Wording

Make sure that the questions are clear, unambiguous and unbiased. Poorly worded questions can lead to biased results. Stay away from leading and loaded questions that prevent you from revealing respondent’s true attitudes and opinions.

Leading questions steer respondents to a desired response.

Example: Do you use ABC brand shampoo?

A better question would be: Which brand of shampoo do you use most often? A list of competing brands should be included for respondents to choose from.

Loaded questions tend to touch on peoples fears and emotions.

Example: Shouldn’t people avoid shopping online to prevent identity theft?

A question worded this way would, of course, result in a “Yes” response as no one wants his or her personal data hacked when shopping online. A better way to get at a person’s unbiased opinion on preventing identity theft when shopping online might be:

Please rate the effectiveness of the following ways to prevent identity theft when shopping online (each way is listed).

Questionnaire Organization and Flow

A survey questionnaire should have a beginning, middle and end. Questions at the beginning of a survey screen for the type(s) of respondent and are followed up by easy to answer questions designed to engage the respondent and raise their interest levels.

Questions in the middle address the survey objectives in detail. These questions tend to be a little more complex and can include rating scales and skip patterns. It is important that these questions come early in the survey so even if a respondent does not finish the survey, their responses can be used in the final analysis.

The end of the survey is where classification questions are presented to the respondents. These questions define and describe your respondents and provide extra meaning to the data during the analysis phase of the research. For consumer surveys, classification questions include demographic information such as age, income, education etc. For B2B surveys, classification questions are about attributes that describe the business, such as number of employees, industry type, company size, location etc. Classification questions are left to the end because some respondents may not answer them because they perceive them as being of a personal nature.

Ten Words to Avoid in Question Development

Certain words force respondents to completely and fully agree or disagree with the question. These words present the respondent with extreme absolutes and should be avoided. As a result, answers to questions with these words may be biased because survey participants could interpret them literally when answering.

Ten Words                                      Source: Burns & Bush, Marketing Research 6th Edition 2010. p. 307

Professional survey researchers know that questionnaire design is a systematic effort that involves the consideration of various question formats, carefully choosing the right wording and organizing the questions in a layout that engages the respondent and minimizes respondent fatigue. However, if you choose to go it alone by following these simple guidelines you will have a winning questionnaire that provides the insights needed to make the best business decisions.

gL Market Research is a full service marketing consulting firm specializing in surveys, focus groups, digital marketing and brand management.

GL Logo w tagline


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s