The Perfect Survey Design: Why Less is More
Buy less, own less, live smaller – between car sharing services, urban bike rental outposts, and the rise of the tiny house movement, minimalism is trending.
Our cultural values may be shifting toward less is more when it comes to consumer goods, but are we becoming more conscious of how we use the most precious commodity of all – time.
Time is precious, whether it’s yours or that of your respondents. When it comes to making the best use of time, survey design needs to adopt a minimalist mindset to prevent survey fatigue.
The concept of survey fatigue is, in its simplest form, the various negative effects on a respondent if they are surveyed too often, or if the survey design itself is poor (lengthiness being a common symptom).
The Rise of Big Data in Marketing
Market research has turned its attention away from small-room focus groups for good reason. Able to reach much larger data sets, surveys have become a priceless data collection tool. The number of companies reaching out to large samples of their customers for survey-based feedback (the ubiquitous NPS for example) has increased exponentially.
Unfortunately, one of the most well-documented mistakes made by marketers, students, researchers, and small business owners alike, is wanting to squeeze respondents for every last drop of data. As a result, a survey design that has tens to hundreds of pages long are arriving in customer inboxes.
When it comes to survey design, think like a minimalist. Less really can achieve more.
Not unlike a 140-character limitation to tweets, if you were to limit yourself to say, seven questions, you would force yourself to focus more on getting as much out of each word possible. The questions themselves would be well-thought out with a format and flow that makes sense to the respondent. Building survey logic into a survey improves completion times and accuracy by serving individual respondents only those questions that pertain to their particular situation.
Tread Lightly When it Comes to Survey Design
Consider the same adapt-ignore-adapt effect we have with ads, but with survey design. And not just based on the volume of instances, but combine it with the multiplier of length.
The longer the survey design and the more often you are asked to fill out long surveys, the less likely you are to participate. And for those that do, the less likely they are to provide unbiased feedback.
In short, it can be really hard to get any sort of constructive or positive feedback from a demographic of annoyed respondents frustrated that their dry cleaner sent them a customer satisfaction survey.
If I scroll through my “Promotional” folder in Gmail, I quickly see feedback requests from a clothing company, an e-retailer (two requests actually), a social networking site, a dentist office, a grocery store chain, and a garden supply store. You can see how quickly (based on the tools that already exist to weed through these messages) we have learned to tune these out.
So what can we do to avoid surveys becoming an annoyance people are keen to avoid? Simply prioritizing the user’s experience to keep surveys as pain free and short as possible will go a long way in reducing fatigue.
How To Keep Your Survey Design Short
This is a great way to make your survey design appear as short as possible, which is a key factor in obtaining more responses. The following question types can be used to ‘trigger’ a follow-up question to appear using Show/Hide triggers:
- Multiple Choice
- Dropdown Menu
- Likert Scale
- Multiple Choice Table
- Table of Checkboxes
After creating at least one question of the above types on the page as the ‘trigger’ and then a follow-up question/text as the ‘target’ that will be shown upon the respondent clicking on a specific answer in the ‘trigger’ question, you are ready to apply the rules.
Show/When logic can be used to conditionally display a question or a page based on the answers to previous questions.
Use Show/When logic when the “target question,” the question you want to conditionally display, is on a subsequent page from the trigger questions. You can also use Show/When logic to conditionally display entire pages within your survey.
Question Piping – Using the reporting value or title of selected answers from a previous question to repeat a question on a later page for each answer selected.
Option Piping – Using option piping you can use the reporting value(s) or titles of selected answer(s) from a previous question as the answer options in a question on a later page.
Test Your Survey Design for UX
Testing your survey for dysfunction before sending is a commonly followed best practice in online surveying. However, testing it for usability is often overlooked. When you create a survey, consider the purpose of each question and whether or not it directly achieves the data collection goals you outlined before starting. Then, when you’ve designed and built the survey, test it for usability. If you’re only halfway through the survey and already tired of clicking radio buttons, chances are your respondents won’t even make it that far.
Consider the loyalty of the respondents as well. If you’re looking for feedback from someone who visited your restaurant once, as opposed to someone who is there each week, you’re far less likely to get thoughtful, time-intensive responses.
Here’s what it boils down to – in order to avoid the externality of surveys becoming the next mutable, fast-forwarded television commercial, do the responsible thing and,
A) Avoid high frequency; soliciting the same sample of people each week won’t get you the actionable data you’re looking for. It may, instead, lose you customers.
B) Keep your surveys short; use the logic tools available and remember the golden rule – if you don’t want to take it, neither will they.