How long should a survey be?

How many questions should a survey have? What is the optimal survey length for high response rates and insightful data? If you’re new to creating surveys, these questions will have no doubt cropped up somewhere along the way.

In reality, there isn’t one simple answer. It depends on the type of survey; its goal and how invested respondents are.

Put yourself in the mind of a survey respondent. How long you’re willing to spend depends on the type of survey it is and how invested you are in the survey, the person who sent it to you, and its results.

It makes sense if you think about it. Say Yorkshire Tea runs a fun, informal survey about what Brits consider a proper brew—always a contentious issue! For someone taking the survey, it might be an enjoyable way to spend a few minutes, and a source of lively debate around the office. But you’re unlikely to want to spend 30 minutes filling it out. From Yorkshire Tea’s perspective, they probably see this more as a branding or customer engagement exercise than as market research. Their goal is therefore to get people talking about their brand and engaging with them. They don’t need the survey to contain lots of questions for rich data. Therefore, this survey is likely to be short with just a few questions.

Next, let’s suppose you’re running the show over at Yorkshire Tea HQ again. Only this time, you want to conduct in-depth market research, which means relatively long surveys. You want to ask potential and current customers about your brand, including brand awareness and brand loyalty. You might have chosen to incentivise people to take your survey, which will mean they’ll be motivated to keep going to the end. Plus, you’re sure to have some lifelong fans in the mix too, who would love the opportunity to give their two pennies’ worth. In this case, survey respondent investment in completing the survey is high, and though the survey will be lengthier, completion rates are likely to remain relatively high. Given this, and the richness of the data you want to collect, you might decide to include some open-ended questions and, here, the optimal number of survey questions would be significantly higher than the previous example.

Another situation where respondents are likely to be highly invested in the survey, is where the outcomes directly impact them. Take for instance a staff satisfaction survey or a 360-degree feedback survey you complete about your line manager. The findings from these surveys will probably affect your working environment, so it’s in your best interests to contribute your thoughts. In fact, SurveyMonkey’s data gurus have looked at survey type and how this affects respondents’ willingness to complete longer surveys. They found that tolerance for lengthier surveys is higher where surveys are related to work or school; and lower when they are customer related.

What all this means is when you’re deciding how long a survey should be, you need to think about the type of survey it is, who your survey respondents are, and how invested they might be in taking it. And this needs to be balanced with the data you want to gather.

How long does a question take to answer anyway? For a 10 question survey, we found that on average, people take just over a minute to answer the first question (including the time spent reading any survey introductions) and spend about five minutes in total. This length of time would work well for our fun Yorkshire Tea proper brew survey.

If a 10 question survey takes five minutes, you might think a 30 minute question would take 15 minutes. Think again! Interestingly we’ve found that the more questions you ask, the less time your respondents spend answering each question. They can start speeding through a survey, as you can see in the table below. This concept is known in the methodology world as “satisficing”. And that can negatively affect your data.

In fact, for surveys longer than 30 questions, the average amount of time respondents spend on each question is nearly half of that compared to surveys with fewer than 30 questions. The other problem with larger numbers of questions is rising survey abandon rates. We’ve found that for surveys that took more than seven or eight minutes to complete, completion rates dropped anywhere from 5% to 20%.

Something else to bear in mind is that some types of questions are quicker and easier to answer than others. Multiple choice questions tend to be the quickest, provided they’re clear and worded well. On the other end of the scale are open-ended questions, which ask people to answer in their own words. This is much more time-consuming and requires more thought. But don’t rule them out right away. For more in-depth surveys or questions where you want to understand someone’s perceptions or behaviour, they can be just the ticket.

Long surveys don’t need to be onerous. You can make use of skip logic to jump to certain questions based on previous answers. Also make sure you read our guide for creating surveys.

A less thorough survey that doesn’t require rich data and has less buy-in from those taking it, like our Yorkshire Tea proper brew example, should have no more than 10 questions.

Meanwhile, a more in-depth survey, like one looking at staff satisfaction, or conducting market research, could be up to 30 questions long, provided respondents are committed to the survey and its outcome. While it can of course be longer, there’s a risk to the reliability of your data as respondents start speeding their way through the survey.

So, how long should a questionnaire be again? Well, when it comes to designing your next survey, you need to balance your audience profile and survey goals with the number and type of questions so you can get the best data possible for the decisions you need to make. In truth, that’s the secret to optimal survey length. And if you do write a survey that has a low response rate, make sure you send it to enough people to get a good sample size.