Survey Design Best Practices And Guidelines


Last updated on July 21, 2023

Surveys are one of the most useful tools for gatherings quick customer feedback and understanding what motivates them. But with great power, comes great responsibility – if you’re going to get results you need to make critical business decisions, it’s important that you’re survey is designed well.

Here, we’ll discuss some best practices for designing effective surveys – everything from creating questions that elicit actionable responses, to logic-checking your data to ensure accuracy. 

So grab a pen (or popcorn) and let’s dive into the world of survey design! 

Carlos Del Corral is an expert on crafting quality surveys. With a track record of working at tech giants like Microsoft and Nokia, he offers top-notch advice to boost response rates while generating deeper insights from survey responses. In this video, Carlos imparts his knowledge gained through extensive experience in market research and product/service design – giving you all the tips needed for successful survey creation!

Survey Design Best Practices

Principles of Survey Design

Designing surveys is an important part of market research, and it requires a great deal of thought and care. It’s essential to keep principles of survey design in mind when constructing questionnaires or polls.

These principles include being aware of respondents’ behaviors, making sure questions are phrased clearly and objectively, ensuring the survey is reliable and relevant to demographic characteristics, as well as minimizing bias.

Additionally, testing survey questions beforehand to ensure they make sense is important before asking respondents to answer them. Those who are responsible for designing the survey should also consider how long a questionnaire should take and whether certain questions are worth asking in light of the target audience.

A well-designed survey has the potential to bring about valuable insights that can be used for future decision-making.

Here are things to keep in mind when creating a survey design:

  • Start with the end goal in mind
  • Simplify
  • Avoid bias and priming
  • Optimize for automated insight generation


Start with the end goal in mind

When it comes to surveying design, the most important principle to keep track of is to start with the end goal in mind.

This means that before you even begin constructing your survey, you should take some time to think through and clearly define what result you want to achieve with this survey.

The goal that you set for your survey will define every aspect of the survey, including its complexity. Here are some things to consider:

  • Who do you need to interview?
  • What kind of results are you trying to get?
  • Are you looking for feedback on a new product or service?
  • Are you trying to understand customer preferences?
  • Are you trying to measure public opinion on a certain issue?

Answering these questions will help guide the design of your survey and ensure that it is tailored toward getting the insights you need.

Simplify to maximize response rate and reduce work

Once you have a well-designed goal, remember to keep things simple when designing your survey. Too many questions or overly complex questions can confuse participants and make them less likely to complete your survey.

Survey fatigue can also be an issue if too much time is required for participants, so keeping your surveys concise and relevant will reduce participant fatigue and increase response rates.

Structure your survey to avoid bias and priming

Avoiding bias and priming is also essential when designing surveys. Bias refers to any question that may lead respondents towards one answer over another, while priming occurs when a participant’s answers are influenced by previous questions or statements in the survey.

Here are 3 simple steps that will help you along the way:

  1. Start with a screener
  2. Continue to generic questions and concrete questions
  3. End with classification questions


Screener – Knowing who should take your survey is essential to getting the results you want. That’s where screener comes in – they help you identify specific demographics so that only those who fit the criteria will respond to your questions. Knowing the demographics for example is a common criterion used for screening surveys – whether it be online or offline market research.

Ask generic questions first and then concrete questions – Start by posing broad questions to understand the overall subject matter before getting into more specific details. This can help ensure meaningful conversations and efficient use of time.

“In Lumoa’s case, if we were interested in knowing more about customers’ overall relationship with the brand, we would first ask about their relationship with us. Only then would we ask about other topics afterward.” Carlos Del Corral

End with the classification questions – These are questions about demographics. For example, age, gender, household income, etc.

These questions are at the end of the survey for two main reasons:

  • By asking classification questions at the beginning, people may wonder “why do they want to know all these things about me?”. It’s better to start with questions that are close to the surface of the things you’re interested in.
  • Second, this type of classification question is usually very clear to the respondent. For example, an age of a person is extremely easy to determine. And when you don’t have to make an effort to answer the question, they’re quick to get answered, and the drop rate will be smaller.


Remember: Being aware of potential biases, avoiding leading questions, and testing out questions on potential respondents prior to launching surveys can help guard against biased results.


Optimizing survey design for insight generation

Finally, optimizing for automated insight generation is key when crafting surveys. The main point here is that in most cases, you can actually go by, or solve with a simple Key Performance Indicator (KPI) and a “Why?” question.

By asking “How satisfied are you with our product?” and “Why?”, you don’t need to ask multiple different questions about the different features of the product. You’ll find out in the “Why?” question.


“This way of asking questions, (“Why?”) hasn’t been used since there haven’t been tools to analyze those open-ended questions properly. But now that you have tools like Lumoa, you don’t need to ask for many concrete things.” – Carlos Del Corral


Utilizing features such as branching logic (showing different sets of questions based on earlier responses), sentiment analysis (measuring how positive or negative customers feel about particular products or services), and open-ended text analysis (allowing respondents to type in their own explanations) can all enable richer data collection and provide more valuable customer experience analytics into customer behavior or public opinion.


Different Types of Questions in a Survey

By now, we know that surveys are one of the most efficient ways to capture accurate data, which can then be used to inform decisions. But having too many (or few) questions on your survey can have a negative impact and make it difficult for you to collect meaningful insights.

That’s why it’s important to carefully consider all the different types of questions you should include in a survey so that you can get reliable and valuable responses from your respondents.

Here are some common types of questions in a survey:

  • Behavioral type of questions
  • Attitudinal type of questions
  • Classification type of questions


Behavioral Questions

Behavioral questions focus on the actual behavior of respondents, such as what they have done in the past or what they’re currently doing. They’re designed to provide information about how people behave. Some examples are:

  • “How often do you visit the doctor?”
  • “How much spread do you buy in a typical week?”
  • “How often do you purchase groceries online?”


NOTE: It would be ideal if you already know your customers’ digital behaviors. For example, how often do they access a particular service (Google Analytics), or how much do they purchase (CRM), etc. Thus, these questions aren’t really needed in your survey.


Attitudinal Questions

Attitudinal questions seek to uncover how people feel about a given topic or product. These questions help researchers understand a respondent’s attitude toward an issue or concept:

  • “How would you rate our product?”
  • “How likely are you to recommend…?” (e.g. typical NPS questions)
  • “Why did you give us that score?” (open-ended questions)


Open-ended questions are important since they allow you to express your feelings, emotions, and sentiments. They’re not just saying that they love your product, but also answering why. This is different from just getting a score of 9 in an NPS rating. Getting insights from the data is easier because it’s richer.


NOTE: Whoever is designing a survey should be careful to ask neutral attitudinal questions that don’t lead in any one direction.


Classification Questions

Classification questions ask respondents to identify themselves in terms of their demographics such as age, gender, race, and other categories. These types of questions help researchers understand their target audience better.


NOTE: In today’s “woke culture” we need to ensure that these questions are phrased in a non-discriminatory manner.


Creating Questions Do’s and Don’ts

Survey designs should always aim to keep surveys concise and easy to understand so that respondents are more likely to complete them. Let’s go through some do’s and don’ts when creating questions.


Avoiding Bias: Symmetrical Scales

The most commonly used scale for surveys is the Likert Scale, which is a symmetric 7-point scale ranging from strongly disagree to strongly agree. Survey designs should have question-wording that reflects this symmetry.

For example, if the question is “I’m satisfied with my purchase”, then the Likert scale should have seven points ranging from “strongly disagree” to “strongly agree” and not just three or four points.


Avoiding Bias: Leading respondents

Leading questions are those that suggest a particular answer to the respondent and will influence their response. Survey designers need to be aware of this and should ensure that their questions are neutral, and framed in a way as to not lead respondents toward a particular answer.


Ask One Question at a Time

One should avoid asking more than one question in the same statement. This can confuse you and your respondents. Leading them to give inaccurate or incomplete responses.


Simple, Specific, and Short Questions

Surveys should be kept simple and on point. Remember to make the questions as easy to read and clear as possible while avoiding the use of complex language or jargon. Survey respondents have limited attention spans and will be more likely to finish the survey if the questions are straightforward and easy to understand.


Avoid Jargon and Acronyms

Following the above note, avoiding jargon and or abbreviations that may confuse respondents should be avoided. If terms have to be used, they must be explained clearly in the survey question itself.


Avoid Questions that Require Respondents to Perform Calculations

Questions that require a respondent to calculate an answer should be avoided as much as possible unless necessary. Because it can lead to inaccurate responses. A workaround would be providing ready options for respondents so that they can select the answer with ease.


Optimize open-ends for rich results

We know already that open-ended questions are a great way for survey designers to get more detailed feedback from respondents, but they can be difficult to analyze. So remember to make sure that the questions are phrased in such a way as to ensure that the response can provide useful insights.


BONUS: Metric Selection – Optimizing for impact within the organization

Selecting metrics and scales can be controversial. On the other hand, remember that there is no one-size-fits-all. Consider what type of business you have and what works for you and everyone who’s involved in it. Let’s quickly discuss the two.

Metric Selection

There are many different metrics to choose from, such as CSAT, NPS, and CES. KPIs can be optimized for different touchpoints, and there’s a lot of theory about how to do that.

Our recommendation is to simplify rather than optimize for marginal improvements. Using fewer metrics and scales, as well as being consistent with those metrics will help make the process as simple as possible.


“As far as metric selection is concerned, Lumoa has a very pragmatic approach to it. We believe that your organization should drive impact in whatever you do.” – Carlos Del Corral

Achieving change within an organization and driving customer-centricity is already challenging, especially in larger organizations. Every time you add a layer of complexity, it becomes more complex.

For example, If you only have NPS, you only need to explain one metric to your stakeholders, or the organization. Adding more to the equation may result in a loss of understanding of what they mean, and practicing becomes more difficult.

Remember these and thank us later:

  • Simplify
  • Be Consistent
  • Reduce the number of metrics in use


NOTE: NPS is not the only metric; there are other widely known/used metrics that can be used instead (NPS is not without its pitfalls)


Scale Selection

Scale selection generates much debate in market research. To achieve consistency,  you should measure across touchpoints and business areas using the same scale, if possible, across all touchpoints.

Just focus on improving, but keep the scale consistent so you can compare and measure improvements.

The scale will be largely determined by the KPI you will choose. Some use standard scales like NPS (1-10).

In selecting a scale, we strongly recommend avoiding creativity. If you do this, it’s impossible to follow the benchmarks. It’s recommended that you stick to the definition of a standard metric or KPI if you choose one.

It may be necessary to force a choice in some cases. Suppose you want to force your customers to tell you if they’re happy or not. If this is the case, a 4-point scale is a good option, since you can have very dissatisfied, dissatisfied, satisfied, and very satisfied. The key again is to be consistent and keep measuring improvement and change.


Generating Insights from Results

With Lumoa, it’s never been easier to gain powerful insights from customer feedback. Whether you’re looking at surveys, transcripts of phone calls and chats, or collecting online reviews – everything can be integrated into one platform!

Forget manual work as all the touchpoints are organized for you in an instant. You’ll also benefit from real-time analytics in over 60+ languages so nothing goes unnoticed. But that’s not all – after gaining your insight, use Lumoa to close the loop by following up on tasks and making sure everyone is held responsible for the voice-of-customer.



Designing surveys is essential for market research but it can be difficult to know where to start. By keeping the principles of survey design in mind, you can create questionnaires that are simple and easy to understand, which will result in more reliable data.

With the help of Lumoa, collecting and analyzing customer feedback from a variety of sources becomes easy. Moreover, you can get the insights you need to make informed decisions about your business.

Now, we can say that you’re definitely ready to start creating your own survey!


Lumoa link

State of AI in CX 2024

Be a pioneer in your industry! Download the State of AI 2024 Report today and join our elite community of over 10,000 CX leaders focused on the future of artificial intelligence.

Latest Articles

Mastering Customer Feedback Strategy: Essential Steps to Enhance Your Business and Service Quality 14 May 2024

Customer feedback is a direct line to how your customers feel about your organization, its products and services. It’s one thing to believe your products are the best in the business, but if they’re not meeting your customer’s needs, you might as well be selling pie in the sky. Customer feedback analytics are invaluable to […]

Read more
Lumoa and GPT: How Lumoa Complements GPT for Actionable Insights 29 Apr 2024

Recently, we did a study at Lumoa on the state of AI in Customer Experience and found out that Artificial Intelligence (AI) has quickly become a core component of improving customer experience (CX), especially in service industries. The uptake of AI in customer experience is very visible in its use across diverse industries, from manufacturing […]

Read more
Medallia vs. Qualtrics vs. Lumoa: A Buyer’s Guide 29 Apr 2024

Every interaction with your business has the power to build or damage the relationship with your customer – and can directly impact your revenue. 78% of customers have backed out of a purchase due to a poor customer experience (CX).  The pressure is rising for businesses to step up their CX game. More than 80% of […]

Read more