This post is about my software product Hyperion. You can read more about it here.
With the first PIT count over, it was time to get customer feedback. I wanted to know:
How much did they care about each of the features? Did they even use them? I only added features that I thought most of my customers would want, but I could have been wrong. If features weren’t being used, then they should be cut.
What big new features should I add to Hyperion? I had ideas from customer interviews, but wanted to know whether those were flukes or trends.
How do my users feel about Hyperion? I used Net Promoter Score, a metric I was familiar with from my time working in tech support.
Are my users planning on renewing for next year? Especially important considering the first year was free.
Would they be interested in writing an endorsement for the website?
Should I have charged the first year instead of making it free? This was just to satisfy my curiosity.
Are there other products that I should build?
What marketing language should I put on the website?
I used Google Forms to make the survey. It was easy.
This is what it looked like:
The Email/Getting Responses
I had working relationships with everyone I was emailing. I had spoken to almost all of them over the phone or in person.
I wanted to stress the following points:
- I would personally appreciate it if they filled out the survey.
- Because it was the first year, it was especially important to get feedback.
- The responses to the survey would influence what I chose to work on.
- There weren’t that many users the first year, and I was trying to get everyone to respond.
I also used a few more tricks:
- I emailed one person directly, rather than many. When you email more than one person you risk the Bystander Effect causing no one to respond.
- I asked a question at the end of the email. If you structure an email as just “letting them know” then it won’t be a mental TODO item. If you ask them a question at the end, then they’ll feel the need to answer you.
- I mentioned a specific and uncommon (and fairly accurate) length of time it would take them to fill out the survey. That way they know what they’re getting into and they know it isn’t going to be a huge slog. If I had said that it would take them “a second,” “a minute,” or “a couple of minutes,” it wouldn’t have told them anything because those expressions are so common that they’re meaningless.
- I time-boxed the request. That’s a good idea in most situations.
- I decided I would follow-up once with people who didn’t fill it out or didn’t respond.
This is the email in full:
I created a short survey to ask about your experience with Hyperion. Because this is the first year, feedback is especially important. There weren’t that many people who used Hyperion so I’m hoping I can get everyone to fill it out. It would mean a lot to me and it will help make the software better (and influence what I focus on).
Here’s the link: [link]
Do you have two minutes to fill it out today or tomorrow?
I have a fairly good relationship my users and most are responsive. I’ll estimate an 80% response rate, so 12 responses (of the 15 users).