Whether you’ve written surveys for decades or are just beginning, the question or survey length is one that’s been around for quite some time, but probably not quite as prevalent an issue as it has become in the past few years. With survey participants increasingly participating via mobile devices, the question of length has become pretty widely debated.
Why length matters
In a previous post on the different types of survey fielding methodologies, I touched a little bit on the fact that the method you choose to field your survey can affect the timing, or length, of your survey. The general rule of thumb is this: don’t make the survey longer than it needs to be.
How’s that for being vague?
But here’s the reality: some surveys will, by nature, need to be longer than others. It will depend on things like your survey’s purpose, your audience, what you intend to do with the data, how you intend to cut the data, and how you’re fielding the survey.
For example, if you’re planning on doing an in-depth study with medical professionals about their interactions with patients, and you intend to do this in person, you will likely have a longer survey than if you decided to get a pulse from your customers about their satisfaction with their latest purchase. In the same vein, if you run a customer satisfaction survey twice a year and do it online, your survey might naturally be longer than if you did a survey with every purchase and expected your customers to answer via mobile.
But how long is too long?
While I could cite studies that tell you precisely how long a survey should and should not be, here are some things to take into consideration:
- About 40% of survey participants are now using a mobile device to take surveys.
- Response rates are decreasing across the board.
- A panel isn’t always the answer to your survey response rate woes.
So, while 20 minutes has been the standard for online surveys, 20 minutes for a mobile survey is simply too long. And with the increase in the number of participants using their mobile devices to take surveys, while I once called for a shift to writing mobile surveys, instead of surveys adapted for mobile, I’m starting to think that taking a device-agnostic approach is better for a whole host of reasons. Among them – not everyone is using a mobile device to access your survey, but not everyone is using a laptop or desktop to access your survey, either. And since you don’t know who is going to access your survey how (unless you already know that particular aspect of your audience already), then creating a device-agnostic survey will be best.
However, more on that later. Back to survey timing.
10 minutes has become the new standard for surveys. I don’t have data to back me up on this, but I’m going to venture a guess that there are extremely few people and groups who meet that standard. And while it’s a great ideal, I still am not convinced that this ideal should apply in all cases. I think that making your survey as direct as possible is a better, more achievable rule. This doesn’t imply that a 45-minute online survey is fine; I believe there are extremely few cases (if any) where a 45-minute survey is a good idea. I have worked on many survey projects, and I can say that even in the most complicated of projects, when you step back and ask, “What am I getting from this question? Do I really, absolutely need this in my survey?” you will find that a good number of questions can be removed while still keeping your intent, focus, and data validation requirements intact.
In other words, the answer to the “how long is too long” question is, as with much else I’ve learned in market research, it depends. Just keep your survey focused, consider all of the variables going into your survey design, and you should have a pretty good shot at having a survey that you – and anyone you know – would be willing to take.