Subscribe to Our Weekly Blog Updates
Get the best in Customer Experience content delivered straight into your inbox.
You have successfully subscribed
Companies ask for feedback so they can get feedback. The challenge is that many surveys seem to be designed not to be answered. Since I joined Medallia three months ago, I have gone on a survey rampage, answering anything I’ve run across. Those annoying intercepts that happen before you even do anything on a website? I answered them. The survey requests at the bottom of receipts? I answered them. I even answered the call I got after my recent car repair.
And I’ve noticed that many surveys seem designed not to be answered. They are long and sometimes irrelevant. I recently received a gift from a brand that I am very loyal to, one that I spend a decent amount on every year. The gift was great, and it came with a packaging slip that included a code I could enter online to respond to a survey. I am on the rampage, so I answered it. But I also abandoned it midway. Why? The third question asked me about my purchase experience. The problem was, I didn’t buy anything. I just got a gift in the mail. The survey didn’t have an “n/a” option and didn’t let me skip the question. I couldn’t lie, so I just abandoned.
So, based on my recent experiences with responding—and on learning about the best practices we teach our clients in our Medallia Institute consulting service on survey design—I came up with a list. It’s my initial list of best practices for survey design to get your surveys answered. Turns out these are best practices that we’ve seen create measurable differences in response rates, so take note.
- Keep surveys short and set expectations up front: Customers are more likely to abandon (or not even start) a survey if they have no idea how long it will take to complete. In your invitation, tell the respondent how long it is and, on average, how long it will take. Under-promise and over-deliver. When the user is responding, show a status bar illustrating how many questions remain. One company found that just by telling customers there were only four questions on a survey, the response rate went up 20%! And the company even got almost half of its respondents to answer more questions after the fourth, which was a much better response rate than for its original longer survey, even though the additional questions added to the length.
- Make surveys mobile ready: If you attended our Mobile Feedback webinar two weeks ago (replay available upon request here) you learned that it is possible to know if a customer logs into your survey on a mobile device and even to detect which kind of device (smartphone, flip phone, BlackBerry, or tablet) he or she uses. With this ability comes the requirement to create a survey that fits the device in both length and layout. Respondents are more likely to answer more questions on a computer. They will answer faster and at a better rate on a smartphone. Make the surveys ready for different devices, and your response rates will go up.
- Send the survey invitation promptly: Statistics show that if you ask for feedback close to the use of a product or completion of a transaction, not only are customers more likely to respond, but their responses are more specific and actionable. If your memory is anything like mine, after too much time goes by, you simply can’t remember the details of your experience. If you can’t remember, you might not respond, and if you do, you won’t be able to say much.
- Give survey respondents the option to provide free-form responses: Scores can only go so far. They tend to miss the “why,” leaving both the company and the respondent wanting more in terms of giving and receiving feedback. Free-form responses give respondents the chance to tell you what they really think and the recipient the chance to get enough insights to take action on the response.
- Ask questions that respondents can actually answer: Back to my earlier point. Make sure to ask questions that are relevant to the respondent. Don’t ask about products they didn’t buy or experiences they didn’t have. Give the option for an “n/a” response.
- Don’t ask for information you already know: If you emailed the survey, don’t ask for the user’s email address. If you know what the customer bought, don’t ask what he or she bought. You get the picture.
Do you have any other best practices for survey design? Let me know! Or respond and let me hook you up with our experts.