Survey Design: The Top 10 Best Practices to Maximize Your Results

Businesses create surveys for a variety of reasons, and good survey design is the key to getting the best results.

Whether you're measuring customer satisfaction or evaluating your new marketing campaign, you'll want to make sure your survey is appealing to your target audience. After all, it doesn't matter how good your survey questions are if no one answers them.

Survey design can make or break your survey's completion rate. Let's review some survey design fundamentals followed by ten best practices you can use when designing your next survey.

As highlighted in the image below, any time you create a survey template, format your questions, or style a survey's look and feel — you're participating in survey design.

But there are some key things to keep in mind during the survey design process. We'll discuss that next.

The Three C's of Survey Design

Whenever you're creating a survey, it's important to consider the three C's outlined in the image below: clear, concise, communicative.

These are the core factors that influence your survey's design and help distinguish between the good surveys and the really great ones.

1. Clear

If your survey design is clear, that means participants aren't left with any questions around the purpose of the survey and how to complete it.

To help determine if you survey is clear enough, you can ask yourself the following questions:

• Are your survey questions easy to understand?
• Does each one elicit a specific answer?
• Will your participants understand the point of each?

If the questions seem hard to understand, random, or out of place, participants are more likely to lose focus — and less likely to complete your form.

2. Concise

Brevity is another big factor when it comes to survey design. Respondents tend to shy away from surveys that seem lengthy or time-consuming.

These four questions can help you decide whether or not your survey is concise:

• How long is your survey?
• Is it the ideal survey length?
• Is it less than 30 questions?
• Are your questions worded succinctly and only asked once (not rephrased or asked in multiple ways)?

Pay close attention to these details as the length of your survey is one of the most important factors that influence its completion rate.

3. Communicative

You also want your survey to communicate your intended message. That way, your results will be meaningful and relevant to your objectives.

Here are a few questions that can help you make sure your survey design is communicative:

• Will the questions help you achieve your main goal?
• Does each one hold weight in producing meaningful insights?
• Are there any questions that are irrelevant or that may distract participants?

Your survey should only include your need-to-have questions, so it's especially important to make the right selection.

The Three C's of Survey Design will help improve the quality of your survey tremendously. But you can also check out the video below from SAGE Publishing to learn more about how you can effectively design the right survey for your audience.

The image below highlights the top ten best practices for survey design.

Let's discuss each one in more detail.

1. Set a goal for your survey.

Before designing your survey, you should come up with a goal or set of goals that you'd like to achieve. Without this benchmark, it's easy to get off-topic and lose sight of your survey's purpose.

Your goal should be simple but specific. Rather than, "I want to evaluate employee satisfaction," consider a more precise goal like, "I want to understand what's causing rapid turnover on our customer-facing teams."

This will provide you with a roadmap to your survey design, making it easier to determine your questions and how best to order them.

2. Strike the right balance between question types.

There are two main types of questions you can choose to include in your survey: close-ended and open-ended.

Close-ended questions are questions that include pre-determined answers created by the survey designers. Typically, these questions come in multiple-choice or checkbox format and participants choose their favorite option from the set of answers provided.

You should include more close-ended questions in your survey because they produce quantitative results. These results are easier for consumers to respond to and for you to analyze.

Open-ended questions can be used to gather qualitative data. But since they take longer to fill out and review, it's best to only include them when looking for specific feedback or when working with smaller audiences.

Be sure to place any open-ended questions towards the end of the survey as they take more effort to complete and can sometimes overwhelm the participant. The best place to put them is about three quarters into the survey before participants experience survey fatigue.

3. Avoid biased and leading questions.

It's easy to accidentally include biased or leading questions in your survey.

For instance, asking, "How wonderful was your experience with our customer service team?" is a common example of a leading question — a question that encourages the researcher's desired response.

Questions like these undermine the validity of your results. You can't trust your data's accuracy because participants have been subjectively influenced by your team.

Instead, you can ask this question by saying, "How would you rate your experience with our customer service team?" This maintains an unbiased attitude, encouraging respondents to answer honestly.

4. Pay attention to your vocabulary and phrasing.

On a similar note, the validity of your data can be jeopardized if your questions are too vague or too limited.

Using absolute words like 'always', 'every', or 'never' forces participants to either completely agree or disagree with your questions. This can make some people hesitant to answer or even complete the survey.

Take this question as an example:

Do you always shop with our company online?

The above question limits responses. Why? Because some customers may shop online occasionally and in-store at other times.

With the answers they're given, it forces them to choose one which decreases the accuracy of their response. Instead, strike the word 'always' in the question and try using a scale that includes options like 'sometimes' and 'rarely'.

Additionally, double-barreled questions — questions that ask participants to respond to two separate sentiments at once — can affect validity as well.

For example, asking, "What do you like best about our website and social media?" may force participants to answer based on their view of either your website or your social media. You won't know exactly which one they're referring to in their answer, and this makes their response relatively useless.

5. Incorporate response scales.

Response scales show the intensity of someone's attitude towards a specific topic. These types of responses provide in-depth feedback on how your audience feels without using open-ended questions.

Rather than offering 'Yes/No' or 'True/False' responses, you can use a 5-point or 7-point Likert scale. This way, participants are presented with a series of statements then asked to rate their opinion using a scale with opposite extremities.

For instance, rather than asking, "Do you come to our stores often?" you could phrase the question like this:

How often do you come to one of our stores?

This format gives you more specific feedback on your topic while maintaining a quantitative, closed-ended response.

6. Keep the wording simple.

Remember the first "C" that stood for "Clear?"

You want to make sure your questions are user-friendly, comprehensible, and leave no room for miscommunication.

A good way to do this is by using casual language with your target audience.

For example:

Instead of this: "What insights did you procure from your conversation with our customer service reps that ultimately impacted your decision to transition from acquisition to advocacy?"

Try this: "How did your customer service experience encourage you to stay loyal to our brand?"

The language in the second version is simple, and every participant will understand it. Avoid the giant run-on sentence.

7. Use images and videos to clarify information.

Sometimes, no matter how well-worded your question is, it still might not be clear to respondents. In these cases, it helps to accompany the question with an image or video to clear up any confusion.

If you want to ask participants how they'd feel about a new product, it may not be enough to describe the concept using just words. Rather than writing a long description, you could include an image for participants to evaluate.

Here's an example question with an image:

Consider the following image before answering the question below.

Image source

How much would you be willing to pay for this smartphone?

This format is much cleaner and easier to comprehend than a block of text. Plus, it also gives you the opportunity to ask a series of questions based on one distinct image.

8. Explain questions around sensitive topics.

For some surveys, you may have to ask questions that could seem unnecessary or personal.

To learn more about your target audience, you'd want to include demographic questions related to ethnicity, income, gender, and more. Some participants are sensitive to these topics, so it's important to explain why you're asking for this information.

If people feel uncomfortable, they might skip your question or worse — abandon the survey altogether. To make it clear why these questions are being asked, provide a short description explaining why they're important to your research.

Let participants know that the responses will be confidential and used only for research purposes. And, of course, follow through on your word.

9. Test your survey.

Once you've designed your survey, you should test it before sending it out to stakeholders. You want to make sure the survey is effective, and that it'll collect the data you're looking for.

Conducting a test-run on a small sample size or internal employees can help your team catch overlooked errors. Getting fresh eyes on the survey will ensure there are no missing questions, misspelled words, or biased wording that you may have missed.

It's best to find these problems now rather than discovering them after the survey's been distributed.

Also, use this opportunity to obtain feedback on the survey's design. Is it too long? Is it boring? Are any questions confusing or repetitive? Do the questions make sense?

Use this feedback to edit your survey, then test it again. Repeat this process until you get a positive response from your participants and are prepared to send out the final draft.

10. Summarize your findings.

So, you've sent out your expertly designed survey. To wrap things up, you'll want to set aside time to review, summarize, and analyze your results.

This is the stage when many businesses begin to recognize the importance of survey design. A poorly designed survey will return, well, poor results.

Designing a survey that's clear, concise, and communicative — and aligned with the other best practices above — will make the reporting process that much easier.

Now that we've covered the best practices, let's take a look at a few examples of survey design.

Survey Design Examples

As shown in the image below, there are several examples of how you can design your survey with different types of questions.

Let's review a few of them here.

Multiple Choice

Multiple choice questions allow respondents to choose one (or more) options from many.

Example (single-option):

What's your favorite time of day?

Example (multiple-selection):

What color(s) would you like to see our product in next? Please select all that apply.

▢ Blue

▢ Purple

▢ Yellow

▢ Red

▢ None of the above

Rank Order

Rank order questions give respondents the opportunity to list different items in their preferred order of importance.

Example:

Which of the following is most important to you when it comes to customer service? Please rank the following items in order of most important (1) to least important (5).

Quick response time ____

Friendly disposition ____

Self-service options ____

Omni-channel support ____

Exceeding expectations ____

Likert Scale

Likert scale questions include statements where the respondent can indicate how much they agree or disagree with each.

Example:

I prefer to shop online rather than in-store.

Matrix

Matrix questions allow you to collect data based on two or more variables for each question or list item you include.

Example:

Based on your recent experience, please rate the following customer service attributes.

Open-Ended

Open-ended questions give respondents the opportunity to leave comments and answer prompts freely, without the constraints of predetermined options.

Example:

How can we improve our customer service? Please type your answer below.

Good survey design is the best way to ensure your survey results are informative and reliable. Make it easy for customers to give you feedback by designing a survey that works best for both of you.

Editor's note: This post was originally published in September 2019 and has been updated for comprehensiveness.