Over the years, a great deal of scientific study and research has gone into the process of creating surveys for research, including B2B research. Studies into the psychological mindset of respondents have yielded key insights into their behavior when faced with certain types of survey questions – for example, the longer a list of responses, the more likely that one of the first choices will be selected. (Some of these insights can be found here, providing guidance in creating surveys.)

But there’s a bit of an art to developing surveys as well. And that art is very akin to acting, because one of the best ways to design an effective survey is to “get into character” as a respondent yourself. By thinking back to surveys you yourself have taken, you can help create more respondent-friendly questions and attract more valuable responses.

The “getting into the respondent’s head” technique is effective in meeting one of the most crucial challenges in developing any survey: motivating the respondents to provide honest answers. This can be tricky, not because people are dishonest, but because they’re busy. Completing surveys takes a certain amount of time and effort which could be devoted to something else on a respondent’s schedule. People need to feel that their time and effort investment will be relatively small and straightforward. Otherwise, people will tend to respond with an intuitive time-saving tactic.

Mailing It In

We’ve all heard this slang expression. It refers to a half-hearted effort that gets a task (especially a tedious one) over as quickly as possible. If survey respondents begin to rush towards the end of a survey, just to put it behind them, their answers will begin to lose value and objectivity. The responses won’t necessarily reflect true opinions, preferences or behavior. “Mailing it in” will save some respondent’s time, but it could end up wasting some of yours. Therefore, keeping the respondent motivated to give good answers is a challenge for both science and art.

The Science

Some scientific approaches to developing more interesting surveys include:

Timing. This factor alone can be the difference between giving good answers, slipshod answers or even not completing the survey at all. Ten minutes may seem appropriate for a survey that you’ve put together to cover all the bases, but that’s a long time in the business world. With some exceptions, 5-6 minutes is as much as respondents will want to invest in a survey (assuming typical questions and knowledge sets by the respondent).

Number of choices. While keeping the number of questions to a minimum is beneficial, the time investment in a survey is really more dependent on the number of choices available. Five questions, with eight options to choose from, can be more tedious than 20 true-false questions.

Length of questions. Generally speaking, questions should fit on a single line on a standard sheet of paper. However, if a question is really important, it’s better to allow slightly greater length than to risk losing the point of the query (also see below).

The Art

These approaches involve more subjectivity and can largely be based on your own experience – creating surveys and, of course, in taking them.

Mix up answer sequences. This is especially important with open-ended “preference” questions involving names for products and services, the timing of newsletters, preferred media for feedback, etc. Researchers have learned that some respondents, when in a hurry, tend to select the last choice on a list. Others, in the same situation, will select the first one. By mixing up the lists at random, you can avoid this.

Mix up question lengths. Longer questions can’t always be avoided, but they should be kept scattered. If your survey contains three 15-word questions and seven questions of less than ten words, try to avoid having two or more of the longer ones consecutively, which can create tedium.

Mix up question types. Experience has shown that the more variation in question types (rating scales, lists, true-false, etc.) the easier the survey will be for respondents. Repeated rating scales (1 through 10; Strongly Agree through Strongly Disagree) can become tiresome quickly for many respondents, who may then begin selecting the same choice repeatedly.

Use “essay” answers sparingly. Use free-association “essay” questions only when it is difficult to capture subjective information, or when the number of possible responses is vast (Example: “What do you like most/least about our Design Workshop sign-in?”). These types of questions obviously take the most time to answer – although in some cases, respondents will appreciate the chance to free-associate.

Through the Respondent’s Eyes

All in all, when considering any B2B research survey, it’s important to take a step back and try to look at the survey from the point of view of someone who’s just received it. Does anything about it seem time-consuming or intimidating? Is the phrasing neutral and objective without being dull and uninspiring? Is it long enough to seem worthwhile without being long enough to make you think of other things you could be doing with your time? Asking questions like these can help you add some artistic touches -- playing the role of a respondent, if you will – to any scientific insights you have learned about creating the best possible B2B survey.