At QuickTapSurvey, we have the opportunity to work with businesses to create surveys that will deliver the type of results they’re looking for. To help you do the same, we’ve outlined ten survey mistakes that can ruin your data. Follow our advice below to maximise the effectiveness of your survey project.
1. Asking too many questions
Asking too many questions is one of the top mistakes you can make with surveys. Users click your survey expecting it will take a few minutes to complete. When they’re still clicking pages later, they’ll either stop filling out your survey or give shorter answers. As a result, you won’t really know what users think. In our experience, 8-10 questions are the ideal survey length.
2. Using the wrong question types
If you ask the wrong type of question, you won’t get the dataset you need. Be sure to define your survey goals and what before you write your survey. Think through what would be most useful to your reporting — multiple choice vs. ranked list, checklist vs. open-ended — when designing the survey.
3. Leading respondents
A survey should tell you what you don’t know, not confirm the beliefs you hold. Remove biased words from your questions so they do not lead users toward a preconceived answer, but allow them to reveal how they think.
4. Loading questions
Don’t assume users exhibit a behavior unless you know it’s true. For instance, you can probably assume all your customers have an email address with online orders. However, you can’t know whether all of your customers have a smartphone. If your survey assumes users have smartphones (say, by asking about mobile websites), you risk getting meaningless data since not all users can truthfully answer.
5. Not allowing customers to clarify
Users often feel pressed to select the closest answer, even if it doesn’t reflect their true feelings. When you allow users to type into a text box to clarify, you’ll get useful information. If you don’t leave space for clarification, you may be able to understand why certain users select certain responses.
6. Combining questions
A survey question should be one question that users can answer, not a multi-part question. Break apart multi-part questions (such as “How do you like x and y feature of our new website?”) to increase survey response quality.
7. Using absolutes
Absolutes such as always/never or ever/never aren’t useful in surveys. Many users don’t “always” or “never” do something, they do it some of the time. A better way to get at user behavior is to ask how often (on a sliding scale) or how many days/week a user does something.
8. Not defining terms
A phrase may be clear for you, but jargon for your users. If you don’t explain what you mean by a word choice, you risk getting irrelevant or wrong answers. Always write your survey using your users’ language — if your clients don’t understand your acronyms, they don’t belong on the survey.
9. Not creating an other option
Letting a respondent select “other” as an option helps users bypass a question that isn’t relevant to their needs. A respondent can also signify that your answers haven’t anticipated their needs by selecting “other.” Paired with a text box for clarification, this can help you learn what respondents want — and where you can capitalize.
10. Always requiring answers
Chances are, your customer base isn’t monolithic. While a topic may be pressing to a third of your audience, ten percent of your users may not share their concern. Requiring every user to answer every question increases the odds of irrelevant data. This can make it more difficult for you to understand user preference and needs.
While this may seem like a lot more work, it pays off big by delivering actionable information you want and need to know. As a result, you can improve your business and best meet customer needs.
Avoiding these common survey mistakes will ensure your data collection efforts get you the information you need to meet your business goals.
Also published on Medium.