A-guide-to-Survey-design.gif

Survey design 

A complete guide to designing a great survey

Table of Contents

 

To get accurate and trustworthy feedback you will need a well designed survey.

This guide is intended to explain the ins and outs of survey design in order to enable you to get the most out of your customer, patient, employee, delegate or student feedback. If you need specific help or would like to talk further about your survey design, or for more information about ViewPoint feedback packages please get in touch.

The basics of survey design

What makes a good survey? Why do some surveys generate an overwhelming number of responses whilst others struggle to engage their audience? How do you ensure feedback accuracy? The overall design of the survey determines the output.

Good survey design focuses on making the survey engaging and appealing to your audience. It avoids unnecessary bias and provides an accessible, logical, easy-to-follow flow. And with so many different methods for running surveys, there are a few important things to think about and some pitfalls to avoid.


Our quick read offers 6 tips for making the customer survey process easy.


Clear objectives

Setting objectives for feedback

Without a clear objective, you won’t know what questions to ask or how best to ask them and you won’t be able to measure the success of your feedback system. The lack of a clear objective will also make it harder for you to gain management and stakeholder buy-in. If you don’t know what success looks like you probably won’t achieve it.

 

Engaging survey design

In a busy world, our minds block out much of what we see which means you typically have a matter of seconds to grab somebody’s attention. A kiosk in a high footfall area, an email landing in an inbox or a comment card at reception, the principal is the same – how will you make sure people engage? The look and feel of the survey is the first thing people will notice. Does it grab attention? Does it stand out from the other things in the environment? How can you use your brand to focus attention? Read more about how to design an engaging survey.

Ask the audience

 

Knowing your audience

You need to understand who you are listening to in order to write a survey that meets their needs. Take a step back and see things through their eyes. Unless you take time to do this the likelihood of you creating a survey that’s going to be engaging and effective is diminished. 5 points to consider are :know your audience infographic

We discuss this in more detail in our blog WHO ARE YOU LISTENING TO?

 

Ensure your audience knows what you want to achieve

Why should your target audience give up time to complete your survey? What’s in it for them? If you give them a good reason for giving you some of their valuable time they will be more likely to engage. Ensure that you position your survey and make it relevant and meaningful to your audience.

Time and context

Be mindful of the time available to your target audience. Delegates at a busy event may have limited dwell time. A short and simple survey is therefore likely to prove most effective in garnering their feedback. However, a survey given to patients waiting in an oncology unit could be longer and more complex as the respondents are likely to have more time to complete a survey and are more likely to be invested in the feedback process than the event delegate.

How to write survey questions

How to write survey questions

A survey sends a message to the respondent as well as gaining their feedback. It sends a message about what’s important to your organisation. It also constitutes an unspoken promise – i.e. if you complete this survey we’ll take your answers into account and we’ll use them to shape our decision making.

What will you ask?

Any two surveys are rarely identical. Different question types and styles can be used to gather different types of data. Read our guide to asking the right questions for different ways to measure satisfaction. The actual questions you ask will be determined by your audience, you objectives and the purpose behind the survey. However some guiding principals are :


"He who asks questions cannot avoid the answers.”

Cameroon Proverb


 

Why? it's important to ask

Why? It's important to ask

Measuring levels of customer satisfaction will tell you how people feel but it won’t necessarily provide insight into WHY customers are satisfied or dissatisfied. To understand this, you will need to see things through their eyes. Why? is such a powerful question in experience surveys!!

Free-text 'verbatim' questions are particularly helpful as they give you the opportunity to hear the unfiltered voice of your customers, providing powerful insights into what matters to them. This is qualitative feedback, where you begin to get underneath how people really feel.

What is your why?

One retailer ran a 3 question in store survey with us, which asked for the overall rating, followed by 6 simple choices for 'why' the customer had given that rating, followed by an open ended question "Is there anything else you'd like to tell us?" We noticed that although the majority of customers gave an excellent rating, 8% of them went on to leave a negative or constructive comment, with a significant number of the comments being about certain shopping experiences. When themed, these experiences were seen to be common to many and adjustments to service, products and ease of buying options lead to an improved customer experience in store.

In another setting there was a marked drop in customer satisfaction on Thursdays. Drilling down in the data showed the time of this was 1 - 2.30pm and drilling down further to a specific Thursday showed that customer comments related to "no one to help in the aisles". By reviewing staff rotas the organisation removed the dip in customer satisfaction and ensured customers could find and buy what they needed. Customer experience improved and sales increased.

Finding out WHY through well written surveys leads to powerful and actionable feedback.

Golden rules for writing a survey

Successful people ask better questions, and as a result, they get better answers.

Tony Robbins


  • Avoid ambiguity - Check your questions to ensure that they’ll be clear to the respondent. Ambiguous questions will lead to unreliable responses. For example, if the answers to a question include “rarely” and “occasionally” as options then there’s ambiguity. What’s the difference between rarely and occasionally? This is subjective and open to inconsistent interpretation. It is far better to replace ambiguous terms such as these with specific ones that remove the need for interpretation.
  • Use complete sentences - In attempting to be brief, avoid the mistake of using incomplete sentences. Whilst this can be tempting, you might be surprised at some of the answers you get from respondents who have clearly misunderstood what’s being asked.
  • Avoid bias in your questions - The question “How would you rate our fantastic new range of coffee?” is an example of question bias. This question prompts the respondent to give a favourable response by describing the new coffee range as “fantastic”.
  • Ensure answers are mutually exclusive - Golden rules for question designSuppose you wanted to gather user demographic data and you included an age-related question in your survey with the options:
  1. 18 -25
  2. 25 – 35
  3. 35 -50
  4. 50+

How does someone who is 25, 35 or 50 respond? Which category should they choose? They are faced with a dilemma which will consume time and may prove frustrating. This can be easily resolved by using a range where the answers are exclusive:

  1. 18 - 24
  2. 25 - 34
  3. 35 – 49
  4. 50+
  • Avoid using negatives in your questions - "Do you support not allowing children in the swimming pool?” is an example of using a negative in a question. The respondent needs to answer “yes” in order to indicate that they don’t want children to be allowed in the swimming pool. Some people will answer “no” by mistake as this intuitively feels like the right answer – i.e. “no, I don’t want children to be allowed in the pool”. In this example, the poor phrasing of the question is likely to lead to unreliable results.
  • Cover all possible answer choices - You could ask the question “Where did you first hear about us?” with the answers:
  1. Radio
  2. Newspaper
  3. Social media

However, this isn’t a complete list. For example, it omits “friend/family”.

  • Avoid skewed answer choices - We recently had a client who wanted to use the following answer options in their survey:
  1. Excellent
  2. Very Good
  3. Good
  4. Fair
  5. Poor

These answer choices are skewed towards the positive. Many people tend to interpret the middle option as neutral regardless of the wording. In the example above, this would lead to results being skewed towards the positive. An appropriate 5-point answer range would be:

  1. Very Good
  2. Good
  3. Neither good nor poor
  4. Poor
  5. Very Poor
  •  Give them a get out! - having a choice to select "none of the above", "not applicable" or "something else" is important to avoid skewed results. Forcing people to answer incorrectly just to get past a question is frustrating for your customer and also can give false representations of the current opinion. For some questions it might be.

Further reading : Ten tips for writing a customer experience survey or questionnaire.


 

Selecting the right question type

Likert scale

Likert-scale-green

The Likert scale is one of the most popular ways of measuring attitudes, behaviours, beliefs and opinions using a range of answer choices on a scale, for example from Extremely Likely to Extremely Unlikely. This enables people to choose from a range of responses rather than expressing an either-or opinion.

Rating scale questions

Rating-scale-questions

This question type asks respondents to rank a range of items or choose from an ordered set. The following is an example of a rating scale question:

When considering a job offer, please rank the importance of the following (Please fill in your rank order using numbers 1 through 6 with 1 being the most important):

Closed-ended questions

closed-questions

A closed-ended question can be answered with a simple ‘yes’ or ‘no’ and is typically used to gather facts about the respondent.

Open-ended questions

open-ended-questions

This question type is used to gain more insight into how the respondent feels. It provides them with the opportunity to provide a free-text response so that you get their feedback in their own words.

Multiple choice questions

multiple-choice-questions

This question type is often used to gather demographic information or to find out about a range of issues. Multiple choice questions can require a single answer or offer multiple answer selections.

 Semantic differential

semantic-differential-questions

Semantic Differential questions are used to ask people to rate a brand, company, product, service, etc. within the frame of a multi-point rating scale. These survey-answering options are pairs of adjectives on opposite ends of the spectrum. For example, happy / sad, love / hate and satisfied / unsatisfied with intermediate options in between to denote strength of feeling, for example:

The Semantic Differential Scale is a reliable way to get information on people’s emotional attitude towards a topic of interest.

Dichotomous questions

Dictotomous-questions

Dichotomous questions are usually used in a survey that asks for a Yes/No, True/False or Agree/Disagree answers. A dichotomous question can only have two possible answers.

 

Which one do you choose?

Selecting the right question type comes down to the type of survey you are managing and how you plan to use the results. A mixture of question types keeps the survey interesting, however too many will look disjointed and could cause confusion. For in the moment surveys we recommend as few questions as possible to keep your audience engaged.

Bringing it together

By improving your survey design you will see an increase in the quantity and quality of your feedback.

A well designed survey will ensure that more people complete all your questions. The greater the response levels, the better the sample size - and therefore, the more informative your reports will be.

Additionally, a well designed survey will ensure that respondents understand your questions within the context of your services, products or their general experience. Where a customer might misunderstand either context or grammar, their responses can be misleading and at worst, totally inaccurate.

A well designed survey or feedback form is the very foundation for a successful, informative and brand-enhancing feedback programme.

We wish you every success - and don't hesitate to get in touch! 


We hope this guide helped you to get the more out of your feedback. If you need any further guidance, we're here to help, just get in touch.


Further reading :

Creating beautiful surveys

Smiley face surveys : Simple ways to see how your customers or employees feel

Capture feedback : Improve satisfaction

GET MORE FEEDBACK.

Do you want help to increase your survey response rates? Get in touch and we will give you the benefit of our years of experience designing smart survey solutions.

 

Get Free Widget
"Such low-cost and yet high-value feedback is immensely helpful in getting closer to our customers. Being able to respond to their expectations will help to keep Shoe Zone ahead of our competition."
Marketing
Marketing, Shoe Zone
"We conducted some research to find out why people leave feedback, and gave options for future feedback systems. Using images, we showed the options available to us. The ViewPoint touchscreen was the overwhelming favourite amongst the staff, students and visitors we asked."
Max Scales
Graduate Project Officer for Student Services Directorate, University of Hull