Customer satisfaction, defined as the degree to which companies are meeting a customer’s expectations for a product or service, always relies on customer input and is driven by customer feedback. Surveys are one of the tried-and-true methods to measure customer satisfaction. When used effectively, customer satisfaction surveys (or CSAT surveys) help companies gain insights about customer needs so they can take action to improve customer experiences.
Developing a survey that addresses business questions may seem easy. However, careful thought and planning must go into each survey element for the results to be useful and actionable. Elements to consider include pre-survey planning, developing the question and scale structure and even creating the look and feel of the survey. Ultimately, designing a sound satisfaction survey instrument helps ensure the data collected through the survey provides solid answers to the business questions at hand.
Consider these 25 best practice tips for creating a valuable customer satisfaction survey that helps you measure customer satisfaction and understand your customers better.
Beginning the survey development process
1. Engage in strategic, thoughtful pre-survey planning.
While strategic planning is an important first step in developing a valuable survey, it’s overlooked by many organizations. Some general principles for survey planning include:
- Target surveys appropriately – Define your audience by determining who you want to obtain feedback from and why. Avoid making surveys too broad by covering too many topics.
- Stay relevant – Make sure every question you ask is relevant to the customer and the customer feedback you’re aiming to gather. Be sure to find out what data already exists and resist asking customers things you already know.
- Be customer centric – Surveys that generate net promoter score (NPS) (asking only how likely the customer is to recommend) are often criticized for revolving solely around what the customer can do for the company and not being attentive to customer needs. Make an effort to meet customers where they are in the customer journey, ensure you’re helping to achieve their objectives, and ask for feedback when, where, and how it makes sense to them.
2. Understand and define the core purpose of the survey.
As part of the pre-survey planning process, determine the information/feedback you want to gain from the survey. Working alongside key stakeholders in your company, you’ll want to ask: What do you need to know from customers in order to take action on the data?
3. Be mindful of survey flow.
To create an unbiased survey, it’s best to think of a funnel when organizing survey questions. This approach structures the survey from general measures to specific measures. For example, asking a respondent about a specific experience with a product or service before asking about the respondent’s overall experience with the product/service, may bias responses to the overall experience measure.
Example of a question to be asked first:
How would you rate the overall quality of our product?
How would you rate our product’s reliability?
4. Keep the survey simple and relatively short.
While it’s never in your best interest to create a long, complex survey, surveys also can be too short. Keeping the survey between 10 and 20 questions is recommended.
5. Include instructions.
While most people today are familiar with how to take an online survey, it’s best not to make the assumption. Include clear instructions for taking the survey. Additional instructions should be included for specific questions, as needed. Some survey elements that may require extra explanation or guidance are:
- Changing a previous answer
- Using drop down menus
- Responding to open-ended questions
6. Explain why.
7. Express gratitude.
It’s important to let respondents know why you’re asking them to complete the survey and how their feedback will be used. The respondent may be more likely to participate in the survey if they know that their feedback will be used to improve a product or service or create a better customer experience. It’s also important to thank respondents for their time in completing the survey. These messages can be communicated in the survey invitation, within the survey itself or both.
8. Group similar questions together.
If there is a list of questions about customer service and a list of questions about product, group those together and avoid intermingling the two sets of questions.
Question and scale development
9. Ask direct questions.
When writing questions, use common, direct language that respondents will easily understand. Customers should not need a dictionary to accurately answer a survey question. Keep questions short and concise. The best survey templates feature sentences that are shorter.
10. Use “outside-in” language.
Write the survey and survey questions with your customers in mind. Avoid using acronyms and jargon, unless you are certain that all respondents are familiar with them and will understand what you mean.
11. Tie every question back to the survey’s core purpose.
Only ask customers what you need to know to meet your business objective. Survey questions should always relate to the survey’s core purpose. Be respectful of your respondent’s time; don’t ask extra questions. “Extra” questions include those for which the answers can be found in a customer database. Any existing demographic information should be pre-populated in online or website-based surveys whenever possible. It also helps to determine whether you can extract an answer from another question. For example, if you ask for a zip code, you don’t need to ask for city and state; those can be identified from the zip code.
12. Refrain from asking redundant questions.
Example of redundant questions:
Customer service is knowledgeable.
Customer service can answer my questions.
From the customer’s viewpoint, these two questions are measuring the same idea. Stay focused on the ultimate business objective by carefully evaluating whether each question uniquely contributes to addressing it.
13. Avoid leading words/questions.
Leading questions are those that guide the respondent in a certain direction. The goal of a good customer satisfaction survey template is to frame questions in a manner that provides objective and meaningful insights.
Example of a leading question:
Our customer service is excellent.
Rephrase to encourage an objective response:
How would you rate our customer service?
14. Avoid double-barreled questions.
Double-barreled questions address more than one subject within the same question and require more than one answer. When drafting questions, be mindful of the word “and” to prevent including more than one idea or concept within a question.
Example of a double-barreled question:
How would you rate the knowledge and friendliness of our customer service representative?
(The representative could have been knowledgeable, but not friendly or vice versa.)
15. Use rating scales whenever possible.
While there are times when asking open-ended questions is important and makes sense, in general, using rating scales is advisable when possible and when it fits with the feedback you’re trying to collect. When developing a survey, considered these important scaling issues:
- What scale will best address the question(s)?
- What type of analyses will be needed to answer the question(s)?
- Given the question and analyses needed, what level of measurement is needed?
- Given the above, how many scale points are needed?
- Response options should be exhaustive. In some cases, this may require including an “other” option.
- Responses to questions should be mutually exclusive; response categories should not overlap one another.
Again, while there are many considerations when developing rating scales, a five-point scale, with the following characteristics, is preferred:
- Fully labeled (text labels rather than numerical)
- Leads with the positive rating first, with this treatment used consistently throughout the survey
- Vertical display
- Scale tailored to the specific question
16. Keep scales consistent.
While this doesn’t mean you have to use the same scale throughout the entire survey, it’s important to maintain some level of consistency with survey questions. Refrain from switching from a 10-point scale to a five-point scale to a two-point scale within the same survey.
17. Ask open-ended questions sparingly and only when necessary.
Asking too many open-ended questions can cause respondent fatigue and may cause customers to abandon the survey. However, there are times when open-ended questions are necessary to gather variable information, details of a particular customer experience or encounter, or personal perspectives.
18. Avoid using grid or matrix questions.
Couching several questions within one question is common but not recommended as a best practice when developing surveys. An example of this is to ask a question like “How satisfied are you with the following:” and then including multiple items for the respondent to rate. Sometimes these lists can be quite long. This approach often leads to “straight lining,” in which the respondent checks the same answer for every sub-question without thinking about each item individually and assigning a separate rating.
19. Don’t require questions to be answered.
Unless there is a prevailing need for users to answer a particular question (perhaps it’s relevant to a key performance indicator or a marketing objective that can only be addressed with feedback from a particular question), requiring respondents to answer questions is not recommended. There are countless reasons why someone might not want to answer a particular question, and there’s a chance the user may drop out or abandon the survey if questions are required.
20. Use a “back” button.
Unless your survey includes complex skips or redirects depending on how a respondent answers a question, we recommend including a “back” button in your survey. This gives users the option of going back to re-read a question or check or change an answer. If you sense that having a back button will confuse or distract a respondent, then consider whether it might be better to leave it out. We tend to offer the same advice for progress bars. It’s often helpful for respondents to know where they are in the course of the survey—and especially when they’re nearing the end. However, as with the back button, if there are redirects or skips that cause the progress bar to jump around a lot, it may be better to leave it out of the survey design.
Survey “look and feel” and final considerations
21. Keep the survey layout simple and clean.
22. Limit format to one question per screen for online surveys.
A simple and clean survey layout will make it easier for respondents to take the survey. Some suggestions for visual survey design include:
- Use consistently formatted answer boxes.
- Limit the length of the question text to decrease the likelihood that a long line of prose will extend beyond the respondent’s visible screen.
- Use dark print for questions and lighter print for answer options.
- Contrast in color has been shown to be more important than the actual colors chosen. More specifically, using lightly shaded colors for the background of questions has been shown to increase response rate.
23. Be mindful of “survey take time.”
The time it takes to complete a survey should be no longer than five minutes for a transactional survey (survey based on a particular point of contact or encounter) and 10-15 minutes for a relationship survey. If you opt to include the survey take time in the survey invitation or instructions, be sure it’s an accurate time estimate.
24. Take the survey for a test drive.
Once you’ve developed the survey, ask some people to take it. Internal stakeholders are one resource, but it’s best to see if you can recruit some customers to take it before it’s fully deployed. Gathering feedback from some actual customers and seeing how they respond to both the survey instrument itself and some of the responses they give can be very helpful in making final modifications to the survey.
25. Don’t skip—or skimp—on pre-survey planning.
This list of recommendations ends right where it started—reinforcing the importance of upfront planning. Pre-survey planning is critical to designing a survey that provides actionable insights your business can use to improve customer experiences. You’ll achieve far better success with surveys when you have a clear idea of the information you want to gather and how you plan to use it.
Start a conversation with Walker Contact Us