Request a Demo

Close Form PopupLink to Close the Form Popup

Demo Form

Learn how we use your personal data in our privacy policy and about our country/region options

Tips for Survey Design and Execution

Written by Alida

Published August 21, 2012

Every year since 2008, Alida has showcased the insight communities that are 'best-in-class' and chooses one recipient for the Insight Community of the Year award. As we gear up for our 2013 competition over the next few weeks, I'm sharing the five key ingredients for a winning community that our past winners - Discovery Communications Banana Republic, NASCAR, MassMutual and Debenhams - have all proudly demonstrated, year after year.

The first three ingredients I've covered in this series are:

  1. aligning your insight community with your business objectives
  2. striving to implement a diverse recruitment strategy
  3. always remembering to engage all members regularly and developing an ongoing retention plan

The fourth key ingredient is closely tied to #3 and focuses on survey design and execution.

#4: Survey Design and Execution

In an earlier post, I mentioned how important it is to design well-written surveys with relevant subject matter, targeting the right members. Let's break these points down further to illustrate the importance for a winning insight community. Well-written surveys are usually designed with the member in mind in addition to meeting the research objective itself. This is often harder than it sounds. You always have the best intentions of designing a survey that will enhance the members' experience; however reality sets in and you're faced with a 50-question interview full of long multi-choice grids about a topic that may only appeal to a small percentage of the entire membership, but you need 600 completes for your internal client by noon tomorrow. Sometimes these situations are unavoidable but wherever possible, try putting yourself in the member's shoes and ask yourself "would I want to complete this?"

In this scenario, you could try breaking the survey into 3 shorter surveys or review the questions and remove anything that seems redundant, or has perhaps already been asked and answered in a previous survey. With insight communities, the power is in the historical data at your fingertips - you don't have to re-ask questions that you already know the answers to. You might be surprised at how many questions you can eliminate just by reviewing what you've already asked. You can also incorporate more engaging question types into your survey design, to make some of the longer exercises a bit less tedious for members. It is also recommended that you remove any disqualification (DQ) points from your survey, if possible. Members have gone through the screening process to join your community and if you are inviting them to take part in a survey, you shouldn't be turning them away if they don't qualify. Ideally you can filter your sample based on previous questions and criteria, but if the topic is related to something you don't know about your members, it's fine to send to everyone: instead of disqualifying people based on certain responses, allow everyone to complete the full survey and filter the responses you need for your analysis.

So that covers what I mean by "well-written" surveys and I've touched on relevant subject matter - use as many previously asked questions as you can to choose the right group. Whilst it's not always possible, in cases where you do have past information to base your sample on, your members will appreciate the fact that you've been paying attention to their responses and will feel like you have made the effort to get to know them individually.

We have found our Insight Community of the Year nominees are trying out different approaches to their surveys than they have in the past, all with the member in mind. These include things like incorporating video clips, getting members to upload images, using visual questions to provide members with a more engaging experience. All our winning insight communities make the most of their question archives, creating crosstabs from key variables and seeing every survey as a way to create a more in-depth profile of each member. Sampling tends to be a mix of specific target groups to the entire membership, depending on the topic and needs for the project. Often these winning insight communities will track which members are being contacted and ensure that anyone who was not targeted for a specific project is sent something else to participate in, for engagement.

Many community managers see the benefit in combining ingredient #3 and #4 by designing short, engaging surveys about fun topics as well as ones required for research, and using the right tone and style to appeal to different target groups. As we move more into designing surveys for mobile devices, we definitely want to create the best experience we can for members - keeping surveys short and relevant, knowing we are dealing with limited time and attention span. This will help to increase response rates and keep members satisfied with the experience.

Our final ingredient for a winning insight community ties all the previous four together: regular health checks. When we talk about "health checks," we mean monitoring on a regular basis the activity in the insight community and the fluctuations in response rates, as well as checking in on the overall member experience to make sure we are delivering on our promises to members. I'll provide more detail on regular insight community health checks in my final post of this series on "The Five Key Ingredients of a Winning Insight Community."