Archive for April, 2009

The Secret of Surveys

April 16th, 2009

At my day job, we have a healthy, if not obsessive dedication towards hearing back from our customers. We also have an unusual statistic: an unusually large number of our customers respond to our email surveys. We don’t offer much in the way in incentives, often times only $25 or $50 dollars in a raffle, yet regardless, 1 in every 10 people we send a request to fills out the survey. Typical industry averages hover at less than 2% response rates

We’ve had several experts tell us beforehand that customers won’t respond to such nominal awards, some say that our users won’t even wipe their nose for the chance at $25. So what’s our secret? We make our users feel like they are a part of something before we ask for the survey.

Secrets to Community Building and Survey Administration

1) We have kept our userbase intentionally small while we perfect our product and build a community. If you offer advice, it will be considered.

2) We respond almost immediately to comments on our blog, twitter @replies or DM’s, and support requests. Support requests are enormously important to respond to, and I’ve had dozens of people tell me we provide the best customer service they have experienced (although they have probably never dealt with Zappos). We use Fogbugz to track support requests, which I highly recommend.

3) We make an effort to actually incorporate user feedback and our respondents know that their voice will be heard. People feel like they have their hand in making this startup a success.

4) We respond to our surveys takers. In our most recent survey, people were legitimately surprised to hear back from us even when we asked if we could contact them for more information!

5) We keep our surveys short. No one like’s their time wasted with 50 question marathons and if someone trusts you enough to click on your link, you better not lose their trust once they get there. We ask 5 or 10 questions max. If you need more than 10 questions, you need a focus group, not a survey.

6) Lastly, make your emails personal. We try to address everyone by their first name when we send email requests. If we don’t have the name, and it can’t be inferred from the email, then we don’t worry about it, but people like the personal touch.

As we scale the business, we hope we maintain this successful feedback rate. We might have to experiment with more focused survey’s on smaller user segments, larger rewards, etc, but we’ll tackle those problems when we get there.