Customer Surveys

People argue for and against using progress bars in online surveys all the time. Personally, I think the length of the survey determines whether a progress bar is a good idea. I should point out here, length of the survey means total questions, not just the longest possible path a respondent may take. Sometimes when people use skip, branch and other advanced logic they forget the total number of questions matters when it comes to the progress bar.

Think about these three scenarios:

Scenario 1: Your cable provider sends you an email survey invitation to complete their customer satisfaction survey. You click through to the survey. At this point, you have no idea how many questions you're in for (unless the email invitation told you), but most likely you were given an estimate of how long the survey should take. You begin answering the survey, because of your responses, you're taken down a specific question path. You begin to wonder how much is left in the survey and notice there's a progress bar! This survey is rather complex on the backend and has several paths and over 120 questions, but any given respondent can only travel down a path of 20-30 questions. If your path was the first path, it will look like you have a lot more pages to go through before the end. At this point you're close to done, but unfortunately, the progress bar makes you think you have many, many more questions left. You bail. You don't have time to complete their long feedback form and will think twice about participating in one of their "short" surveys in the future.

Scenario 2: Your cable provider sends you an email asking you to complete their customer service questionnaire. You click through to the survey, answer several questions and notice the progress bar at the top has hardly moved at all. You begin to think, "uuuuuugh this was a mistake!" Then suddenly the progress bar shows you're almost done. Because you're curious, you go back and change your answers to see what other questions the cable provider is asking customers (after all you may want to add your two cents). Suddenly, without meaning to you've messed up their data by not respondent honestly.

Scenario 3: Your cable provider sends you an email invitation to complete their market research survey. You click through to the survey. You answer several questions begin to wonder if the survey will be over soon, but you keep going. A few questions later, you've finished the survey in the amount of time the cable provider's email marketing said you would. Even though it was a little annoying to complete the survey, it was a good experience, and hopefully it will improve your service! Furthermore, they kept their promise on the time and haven't lost your trust.

If you're conducting a business survey, whether it's an employee feedback questionnaire, course evaluation or client survey, which of the above scenarios would you like your respondents to experience? I'm hoping you're thinking to yourself "Scenario 3, of course!" I certainly want my respondents to finish (not abandon) my survey and give me honest feedback (not change their answers to see what else I'm asking).

That said, progress bars wont discourage people from finishing a short survey with honest answers. As part of your survey design, you should consider whether a progress bar will add to the respondent experience or take away from it (and your results).