Fielding Your Survey
Episode #6 of the course How to conduct a market research survey by Nick Freiling
We just learned how to find your ideal survey audience. Today, I’ll explain the right way to get them to actually take your survey.
Market research surveys are pointless without a lot of high-quality responses—feedback from real people with real opinions about your product idea. Here are three things to keep in mind after you’ve finished designing your survey and have identified a way to get answers from the right people.
It’s important to test your survey before launching. No matter how carefully you programmed your survey instrument, you’ll inevitably make a mistake here and there that can cause huge problems down the road. So, take it yourself five or six times. Go down every logic “branch” you set up, and look at each question from the perspective of your respondents.
Sometimes what seems obvious to you may be confusing for your respondents. Remember, they’re just taking a survey—they don’t know anything about you or your product and may not resemble the type of consumer you have in mind.
Some platforms allow you to run “test responses” en masse—do that! This means you’re able to simulate 1,000 (or whatever number) of people taking your survey, then go look at the “Reporting” or “Analysis” tab to be sure every question is being answered by the right people, based on your logic branches.
If you’re sending your survey out yourself, writing a good introduction can make or break your response rate. The rule of thumb here is short and sweet … and intriguing.
The fact is, people get too many emails. More specifically, they get too many requests to take surveys. In order to break through this cloud of noise, your survey’s invitation needs to be perfect. What does that mean?
First, the subject line should be intriguing. Avoid subject lines like “Take my survey” or “Your feedback needed.” Instead, write something like, “Your thoughts about a new product,” or, “This new product made me think of you”—something that gives your recipients a reason to want to click on the email to see what this is about.
Then, keep the body of your email short. Briefly explain what you’re doing and why you want this particular person’s feedback. If you have the budget for it, offer a gift card sweepstakes and tell recipients the value (and their odds). And at the very least, make sure everyone who sees the survey knows where to go with any questions or thoughts.
I’m seeking feedback on a product idea. Knowing your background, I thought you’d have a helpful perspective.
I’m offering a $100 giftcard to one of this survey’s respondents. I expect about 100 responses, so your odds aren’t bad!
Here’s the survey link: www.survey.com/mysurvey.
Thanks in advance!
Finally, bold your survey link. Make sure it’s totally obvious how one can start your survey.
This means sending your survey to 15% of your total sample. Say that you want to survey 300 people—send the survey first to 45 people (15% of 300), then look at your dataset in the “Reporting” or “Analysis” tab. Make sure things look like you expect before sending your survey to the rest of your list.
If you aren’t the one sending your survey to your audience (i.e., you’re using a panel provider), be sure to ask them to soft-launch before opening it to everyone. This should be a standard practice.
Once you’ve ensured that your data looks like it should, open the survey to everyone who fits your targeting criteria. Remember, try to leave half of your responses open to people who don’t necessarily fit your target market but might have helpful thoughts and feedback.
At some point, your survey is going to “close.” This means one of two things: You’ve gotten as many responses as you’ve paid for (if you’re working with a panel provider), or you don’t think anyone else is going to take your survey (if you’re sending it out yourself).
At this point, you’ll need to clean your survey data—we’ll discuss that next.
Founders at Work: Stories of Startups’ Early Days by Jessica Livingston
Share with friends