Which of the following best describes your ability to write a survey?

Poor
Somewhat good
Good
Very Good
Excellent
I don’t know

I am guessing you selected at least ‘somewhat good’ in response to the above question. Chances are at some point in your life you have asked other people questions about their opinion on something, taken an informal poll, or even a written and fielded an actual survey before. So it’s likely you could go about writing a survey right now if you had to.

But what about a great user survey? One that yields insights leading to a change in direction for an entire user experience or business? Well, that’s a bit trickier. For user experience professionals, crafting a survey that effectively unveils deep insights about their users is critical to product success.

Here a few techniques I use to make my user surveys as insightful as possible.

1. Invite and inspire users to respond.

It’s hard to argue with the fact that the most important aspect of fielding a user survey is getting actual users to respond. I suggest encouraging your users by showing them you value their time and effort. Everyone likes to feel valued, right? A good route is to invite users to participate in a ‘survey panel’ and providing an incentive. Inviting users doesn’t just let them be in control of the communications they receive from you, but allows you to create an actual sampling strategy as well (such as random sampling) and collect feedback over time. Inspire your survey takers to provide quality feedback and get more responses by offering them something in small return, such as discounts, gift cards, or other small tokens of your affection.

2. Catch users in the act.

The other most important aspect of a user survey is to gather feedback on actual use of the app, website, or service. But since you’re doing a survey you can’t actually be there in person with the user or determine when the use takes place. Instead you’ll have to settle for the next best thing and intercept users to answer your questions during, soon afterward, or as close as possible to the actual use act while the experience is still “fresh” in their mind. Some great tools that can help you do this include Ethnio, dScout, and texting links to your online survey so users can respond while they are out in the world using their mobile phones (I use Survey Monkey as it works relatively well on the iPhone).

3. Create context.

Related to making sure your questions are responded to (as much as possible) in the mindset of actual use, start out questions by describing a realistic scenario or situation that can put users in the mind set, leading to a more authentic evaluation. For example, start your question with "Think about the last time you looked at..." or "From when you first woke up and started your day today, how often did you..."

4. Focus on multiple choice and ratings questions.

Yes/no questions are a total dead end and often the most leading questions in surveys. Focus instead on multiple choice and rating questions that get at user behaviors, not opinions. Include "other" choices and "please explain" options to make sure you don't miss any insights. Limit any essay (free form) questions to just one or two, although it will be tempting to add more. If you need more in depth qualitative feedback, choose to do user interviews instead.

5. Make it short.

Yup, exactly that. Take all the questions you want to ask and get rid of most of them. Ensure the survey takes no more than five minutes to complete. Yes, I know it’s painful, but the shorter the survey, the more engaged your responders will be, and the more accurate their responses. You’d rather have better data than more data, right?

Finally, once you’ve got your new more-awesome-than-ever survey drafted, make sure you test it out on a coworker, family member, or friend to make sure the questions are clear and to catch any typos. Then put it out here in the world and show your users how much you care about their experience and feedback.

Good luck!