If you’ve ever had the chance to talk to me about survey design, or if you’ve read any my posts about the topic, one of the primary items I refer to is this: start with the one question you want to have answered by your study. Why is this so important?
Single focus makes for a better respondent experience
Let me give you an example of a survey that I took that obviously had multiple questions they wanted to address via the survey. This survey ended up being a little over 20 minutes long. So, it was long, but it didn’t quite break the “way too long” rule. However, the first part of the survey started something like this: “First, we’re going to ask you about your experiences shopping recently.” Great! I can do that. I was a little nervous by reading the word, “First,” but okay. Then, after the questions about shopping experiences, I saw this, “Next, we’re going to ask you about your experiences buying auto insurance.”
Hold up. How does that have to do with shopping experiences? Are they trying to find if I approach retail and auto insurance similarly?
Once that question set was done, though, there was another set: “Now, we’re going to ask you about vacations.”
To me, it felt like three surveys had been included in one, because each section of this survey was taking about eight minutes for me to complete. Each section felt pretty thorough, too. From asking about brands I was aware of, to how often I purchased each brand (retail), to when did I last change auto insurance carriers and who did I get quotes from, to how often and where did I take vacations. It really felt like each survey was a stand-alone survey. I couldn’t figure out how the three sections were being tied together.
A second survey I took started with something like this, “While the questions you are going to be asked might not seem to relate to each other, we’re interested in understanding your brand awareness across multiple categories.”
That survey, instead of diving deep into each category, asked a few similar questions for each category: what brands were top of mind for that category; what brands when presented with a list did I recognize; where did I remember hearing about those brands; when was the last time I purchased from any of those brands. Overall, it stayed true to the original intent listed, and the result was a far more cohesive experience.
The data say…
Another benefit to keeping your questionnaires targeted is that it’s far easier to draw conclusions from the results. Let’s say you went a little crazy with your last survey for your business and asked about the following:
- Customer’s first reactions walking into your physical store
- Customer’s first reactions going to your store’s website
- Customer’s interactions on your social media sites
- Customer’s purchase experience in the physical store
- Customer’s purchase experience on your store’s website
- Customer’s experience with your customer service team
- Customer’s likelihood to recommend your company
When it comes to analyzing the results, you might want to try to see if there is some correlation between their experiences in your physical store, your online store, and your social media sites and their likelihood to recommend your company. However, because you also asked about purchase experience, you’ll need to also take that into account. And because you asked about their experience with your customer service team, you’ll need to include that into your analysis, too. So now you’re looking at six different variables that all have an effect on the likelihood that customer would recommend your company. Are you sure you’re drawing the right conclusions? Not only that, but if this survey ended up being over 20 minutes to complete, are you sure that the customer answering the study isn’t now impacted by the fact the survey took awhile to complete, and perhaps their likelihood to recommend your company is being affected by the length of the survey?
Here’s a better approach, if you still want to get all of that information.
First, start with the following in your survey:
- Customer’s first reactions walking into your physical store
- Customer’s purchase experience in the physical store
- If they visited any of your social media sites, their experience with those
- Customer’s likelihood to recommend your physical store
Survey two would look like this:
- Customer’s first reactions visiting your store’s website
- If they visited any of your social media sites, their experience with those
- Customer’s purchase experience on your store’s website
- If they needed to contact customer service, their experience with the customer service team
The benefit here is you can target your audiences with these surveys. You can encourage customers that do visit your physical store to take that survey, and then also encourage visitors of your website to take that survey. By keeping each of the sections listed to two or three questions per section, you have a pretty succinct survey (and using logic, are asking questions relevant to each customer).
What if research results in more questions?
Do a follow-up study.
Often, research done can result in more questions. Let’s take your physical store survey as an example here. For the “first reactions visiting your physical store,” you want to keep the original survey pretty high-level. Did they like the layout? Did it seem inviting? Did someone greet them when they walked in? That would be the extent of the questions I’d recommend to keep it high level.
If the results come back from this survey showing that customers aren’t liking the layout, then you can follow up with a couple of options: either a more in-depth survey asking specifically about the layout, or you can do in-depth interviews with a select number of respondents (you could ask in the first survey if they would be all right being contacted).
Be sure to maintain the same mantra for the follow-up studies: keep it focused on one objective.
