Engagement surveys Research

Why short surveys get better answers: what the research says

Angelina Korolchak, CEO, Org360
, 6 min read

Org360 surveys have 12 or 24 questions and take about 5 or 10 minutes. That limit is a design choice. Here is the research behind it, including the places where the evidence is mixed.

1. Fewer people start a survey that looks long

In a well-known experiment, Mirta Galesic and Michael Bosnjak told people a web survey would take 10, 20 or 30 minutes. The longer the stated length, the fewer people started it and the fewer finished it. About 75% started the 10-minute version, 65% the 20-minute one and 62% the 30-minute one.

Real-world data points the same way. SurveyMonkey researchers analyzed 25,080 surveys run on the company's own panel and found that completion rates fall as surveys get longer: an average of 89% for 10 questions and 79% for 40.

A 10-point drop sounds small. In a team survey it isn't, because the people who leave aren't random: they're often the busiest or the least convinced the survey will change anything, which are exactly the views you need.

2. The people who stay give thinner answers toward the end

Dropping out is the visible cost. The hidden one is what happens to the answers of people who keep going. In the same Galesic and Bosnjak experiment, questions placed later in the survey were answered faster, with shorter open answers and more uniform ratings than the same questions placed early.

Survey researchers call this satisficing, a term Jon Krosnick borrowed from Herbert Simon: giving an answer that seems good enough instead of thinking it through. A tired respondent ticks the same box down a page of questions. In the report, that answer looks exactly like a considered one.

3. Open questions cost the most, and they're worth it

Writing an answer is harder than picking one. In SurveyMonkey's data, surveys with 10 open questions had an average completion rate of 78%, against 88% for surveys with one. Surveys that opened with an open question also finished less often than surveys that opened with a simple choice.

Open answers are also where the most useful feedback usually is. So Org360 surveys keep 1 to 3 open questions and put them at the end, after the rated ones.

4. Knowing where the end is helps, though a progress bar doesn't do the job

People push harder as a goal gets closer. Clark Hull first described this "goal gradient" in 1932, with rats running faster as they neared food. In 2006, Ran Kivetz, Oleg Urminsky and Yuhuang Zheng showed the same pattern in people: café customers bought coffee more often as their loyalty card filled up, and customers given a 12-stamp card with 2 stamps already on it finished faster than those given a blank 10-stamp card.

You might expect a progress bar to use this. The evidence says it isn't that simple. Frederick Conrad and colleagues found that when a progress bar showed slow early progress, more people gave up. A meta-analysis by Ana Villar, Mario Callegaro and Yongwei Yang found that progress bars don't significantly reduce drop-offs overall, and the SurveyMonkey study found higher completion in surveys without them.

What does help is honesty up front. Tell people the real length in the invitation, and keep the survey short enough that the promise is easy to keep. "12 questions, about 5 minutes" is a promise people believe.

5. A simple, familiar scale

In their review of questionnaire design, Jon Krosnick and Stanley Presser recommend rating scales of about 5 to 7 points, with every point labeled in words. Org360's rated questions use a 1 to 5 agreement scale, from strongly disagree to strongly agree, so people learn it on the first question and can then focus on the content.

Where the evidence is mixed

Shorter isn't automatically better. In a 2021 experiment in Austria, a 35-minute version of the European Social Survey produced only a slightly higher response rate than the 50-minute version, and no clear gain in data quality. Reviews of the research also find that topic and motivation can matter as much as length.

We think this fits workplace feedback rather than contradicting it. People answer a long survey carefully when they care about the topic and trust what happens next. At work, both are fragile. A short survey is one way to show you respect people's time, and acting on what they said is the other. Our feedback guide covers both.

What this means in practice

  • Use about 12 questions for a regular pulse and about 24 for a full picture.
  • Keep 1 to 3 open questions, and put them at the end.
  • Say how long it takes in the invitation, and make sure it's true.
  • Ask only questions you're prepared to act on. Every extra question costs answers somewhere else.

Sources

  1. Galesic, M., & Bosnjak, M. (2009). Effects of questionnaire length on participation and indicators of response quality in a web survey. Public Opinion Quarterly, 73(2), 349-360. doi.org/10.1093/poq/nfp031. Start rates as reported in Survey Research Methods (2025), below.
  2. Liu, M., & Wronski, L. (2018). Examining completion rates in web surveys via over 25,000 real-world surveys. Social Science Computer Review, 36(1), 116-124. doi.org/10.1177/0894439317695581
  3. SurveyMonkey. Tips for increasing survey completion rates (analysis of more than 25,000 SurveyMonkey Audience surveys). surveymonkey.com
  4. Krosnick, J. A. (1991). Response strategies for coping with the cognitive demands of attitude measures in surveys. Applied Cognitive Psychology, 5(3), 213-236.
  5. Krosnick, J. A., & Presser, S. (2010). Question and questionnaire design. In Handbook of Survey Research (2nd ed.), 263-314. PDF
  6. Hull, C. L. (1932). The goal-gradient hypothesis and maze learning. Psychological Review, 39(1), 25-43.
  7. Kivetz, R., Urminsky, O., & Zheng, Y. (2006). The goal-gradient hypothesis resurrected. Journal of Marketing Research, 43(1), 39-58. PDF
  8. Conrad, F. G., Couper, M. P., Tourangeau, R., & Peytchev, A. (2010). The impact of progress indicators on task completion. Interacting with Computers, 22(5), 417-427. Open copy (PubMed Central)
  9. Villar, A., Callegaro, M., & Yang, Y. (2013). Where am I? A meta-analysis of experiments on the effects of progress indicators for web surveys. Social Science Computer Review, 31(6), 744-762. Open copy (City, University of London)
  10. Surely shorter is better? A questionnaire length experiment in a self-completion survey (2025). Survey Research Methods, 19(2), 139-152. doi.org/10.18148/srm/2025.v19i2.8348

Get new research by email

About one email a month, when we publish something new on coaching, 360 feedback and engagement surveys.

We'll email you once to confirm. Unsubscribe anytime. See our data protection policy.

Run your first survey free

12 or 24 questions, about 5 or 10 minutes, and a report you can act on.