As user experience experts, it’s natural that we at Toolhouse are big proponents of testing the experiences that we craft. Whenever possible, we try to maximize our clients’ budgets and schedules to include multiple rounds of usability tests. On occasion, however, our desire for validated user feedback doesn't align with the budget. When recruiting users for interviews isn’t an option, we rely on that tried-and true-conversation starter: the survey.
I’ll cut to the chase: surveys are usability's redheaded stepchild. Marketers never claim to love them; respondents often loathe them; but they are truly a necessary evil that can yield beneficial results for both sides.
It’s really hard to beat the benefits gained by looking over a new user’s shoulder in the lab. However, few marketers have the time or budget to test regularly enough to truly understand how people are engaging with their web experience. The next best thing is to keep an open dialogue with your users. This can be easily executed through web surveys.
There are endless surveys that violate their respondents on so many levels that the quality of the survey results are inaccurate or unusable. Of the hundreds of surveys I’ve written, the most successful resulted in clear, actionable feedback that could be quantified to support business decisions. Here are my top five recommendations for writing a successful web survey.
1. Be strategic and identify 3-5 things you need to know. Before putting pen to paper and rattling off the 20 burning questions you have for your users, be strategic. Identify and prioritize a small handful of things you want to uncover as a result of this survey. Respondents will appreciate you for asking direct questions and they’ll feel respected if they understand your intention. For example: you may need to know if users are satisfied with the new “Find a Provider” tool, or you might want to confirm that doctors are able to find dosing information since you recently redesigned your navigation. Keep your questions focused on those topics.
2. Invite users to take your survey and set their expectations. Numerous studies have proven that cleaner data comes from surveys that not only invite the user to volunteer their feedback but that also take the time to set expectations. It’s as simple as saying, “We’ve made updates to our site, and we value your feedback. Would you please take three minutes to give us your input?”
3. Ask mostly close-ended questions. To do so, you’ll need to have a deep understanding of the experience you’re asking about. You can save time by simply asking "What did you think of our new ‘Find a Provider’ tool,” but your results will be much more difficult to analyze than if you ask a series of questions. Rather than the single catch-all question, it would be best to ask “Did you use the ‘Find a Provider’ tool today?” This allows you to qualify the user; you can now narrow your results to only look at feedback from respondents who used the tool. Then, ask follow-up questions to the audience that used the tool: “Did you find the tool helpful?” “Have you contacted a provider you found using the tool?" While feedback resulting from open-ended questions can be interesting to digest, it’s not easy to quantify. It often requires the analyst to interpret the feedback, which can introduce bias to the results, and makes it more difficult for you to make data-driven business decisions.
4. Always include one final open-ended question to allow your respondents to “vent” or provide additional feedback. For the very reason you shouldn’t use only open-ended questions, you do want to include one that can add some “color” to your results. Open-ended questions beg for personal rants and opinions. They also give the user an opportunity to share something positive with you about their experience that you may not have asked about. If the majority of your results are quantifiable, a quick cross-tab of those results with respondent's open-ended feedback can net extremely valuable insights. For example, an HCP comes to your site looking for co-pay cards and when asked if they found everything they were looking for, they select yes, so you could easily assume that they had a successful visit. But when asked, "If you could make one improvement to the experience, what would it be?” the HCP states that they wish they could share a link to the co-pay cards page with their patients. This presents a great case for exploring options to add functionality to your site. In this example the feedback uncovered in the open-end answer provided added value.
5. Do something with all of that data. Remember how I said this was an opportunity to have an open dialogue with your users? They want to see your website evolve and improve. If users start to notice that their feedback is making a difference, they'll continue to provide honest feedback the next time you ask for their opinion.
Stop the survey bashing already and use one to start a dialogue with your users today. Once the results start rolling in, you’ll be glad you did