[derek_frome]

You'd be amazed. I was an account manager at Medallia (Qualtrics competitor, also backed by Sequoia) and the response rates on 20-50 question surveys was typically in the double digits. When you have hundreds of thousands of customers in a month, you can generate super rich data through this process.

However, the industry as a whole has been coping with something they call "survey fatigue" which is reflected in a lot of the comments here. There has been a general backslide in survey response rates, though as of a few years ago, it had mostly stabilized around a new normal that was still plenty good for generating excellent data sets to understand the customer experience.

[njarboe]

The customer that fills out a survey must be a very select group of customer. If the response rate is in the single digits of percent, what about the rest of the people? Setting company policy on just this small fraction of customers could be the path to failure.

[covemarkets]

correct, and that's why surveys need to be as scientific as possible in order to avoid various biases that skew the results. hence qualtrics and not surveymonkey for serious survey research.

[njarboe]

How does Qualtrics get people who refuse to answer surveys to do so? Come to their door and offer big bucks and then weight those answers heavily?

[covemarkets]

I don't know, but the key is to properly account for the biases in responses of those who do answer. Substantial percentages of people do answer surveys. Don't make the mistake of thinking along the lines of "i would never answer one, therefore nobody would"