UX Survey it's a great UX Research tool to gather meaningful data to support design decisions.
Since UX Design is all about developing products to provide a better experience to the user—and not ourselves—there's no room for personal opinions or bias. And that is why research is so important.
UX Survey is one of the many methods that help researchers collect data.
This article will explain UX Survey and how to create one, together with some tips!
What is UX Survey?
UX Survey is a fast way to gather data and information for a UX Research process. It is basically a form with questions that users will respond to.
The most important step in creating a survey is to develop the right questions. If you don't choose your questions wisely, they won't provide you with relevant data or insights. Ultimately, it can even lead to bad decisions.
UX Survey is not only an efficient and practical tool but also very flexible. Surveys allow gathering qualitative or quantitative data because you can build both open and closed questions, which we'll see more about it later.
Qualitative and quantitative research
Qualitative and quantitative are two types of research whose differences are related to the kind of data and information they collect.
Qualitative research gathers data to study people's behavior, habits, opinions, and preferences. Thus, qualitative results are more abstract and challenging to analyze, requiring more time to work on.
On the other hand, quantitative research aims to collect measurable and comparable data. Quantitative analysis gathers and works with numbers, statistics, and mathematical probabilities. Thus, this type of research is faster to work with because its results are precise.
Because of these differences, UX researchers may feel more inclined to run qualitative research, considering it helps understand user behavior—a crucial step for product development and usability.
However, quantitative researches are also very useful to UX because it can:
- add numbers into the analysis, making it easy to build an argument for supporting decisions;
- allow comparison, between old results and new ones, for example;
- clarify risks and benefits involved in each decision;
- alignment of business objectives with UX Design decisions.
Therefore, it's good that researchers choose both qualitative and quantitative methods as they complement each other and add insights to the decision-making process.
When should you run a UX Survey?
UX Survey is a basic tool among several other research techniques in UX Research—still, extremely important. Thus, it is essential to know when you should make one.
In general, UX Survey will help researchers to:
- find out why users visit a website or download an app;
- collect quantitative data to validate the results of other research;
- provide feedback about a new feature or version of a product;
- gather quantitative data regarding the quality of a product;
- measure users' satisfaction.
On the other hand, UX Survey may not be so helpful to:
- understand users' behavior — field research or usability testing is a better fit for these situations;
- find usability issues;
- unravel users' needs and pains.
Therefore, consider your objectives and what you need to discover before running a UX Survey.
Hands-on: Creating a UX Survey
One of the most dangerous pitfalls researchers can fall for is believing that creating a survey is simple.
Actually, a UX Survey needs careful preparation; otherwise, the data collected will be irrelevant and the team will only waste time.
So, check a detailed step by step to create your UX Survey:
- Problem discovery;
- Prepare the questions;
- Recruit participants;
- Analyze the results
1) Problem discovery
The very first step in preparing an efficient UX Survey is to know what is the problem you want to analyze and solve.
This understanding will guide you through writing the right questions and analyzing the results.
Focus on understanding the research objectives and be sure that a UX Survey is the best way to find answers to that problem. The following situations are good examples to create a UX Survey:
- Is the performance of the app's new version better than the older one?
- What is the satisfaction level of the users with our product?
- Which is the most used feature? And which is the less used one?
2) Prepare the questions
Preparing the questions is the most critical step of the process. The research problem and objectives should be the guiding point to writing them down.
Wrong questions will lead to inconsistent results and bad decisions. In other words, neglecting this phase will make the team waste time and money because what you ask must be relevant to the answers you need.
Therefore, don't skip or underestimate this step to ensure the efficiency of the survey.
Types of questions
One thing that helps you in this process is knowing that there are two types of questions:
- Close questions: the respondents have a limited number of answers: Yes/No, multiple-choice, or Likert scale;
- Open questions: the respondents are free to answer how they like; there are no pre-defined answers or limits.
Open questions are suitable for gathering qualitative data, while close questions are used to collect quantitative data.
Regarding the analysis of results, open questions require more effort and time to evaluate, while close questions are faster to analyze.
Similarly, close questions tend to have a better answer rate than open questions—just because it's more practical to answer them. So, also take this into account when preparing your questionnaire.
A few tips for writing good questions
Ideally, the questions should guide the user and make them answer with no biases that could affect the final results.
In this sense, when writing the questions, try to:
- be as neutral as possible;
- make one question at a time;
- avoid asking the same thing repeatedly;
- avoid questions about behaviors (run a usability test instead);
- use balanced scales answers (when using the Likert scale);
- have neutral answers options like "I don't know" or "Others";
- avoid biased questions.
Examples of badly written questions:
- what did you hate the most in this interaction? (suggestive/bias);
- what do you think about this new product design and its new features? (multiple questions at once);
- how did you get on this page? (behavior);
- how do you evaluate the navigation experience? how do you evaluate the buying experience? (similar questions repeatedly).
3) Recruit participants
A UX Survey is not helpful if there are not enough responses. So, recruiting participants is fundamental to gathering information about your product and its users.
If you have already run research, you probably have a list of potential participants. If you don't have an existing list, you will need to build one.
Remember that the participants must be existing customers of your product or fit your user persona profile. There's no use in surveying people that are not your target audience.
How to make people respond to a UX Survey?
Although a survey is an excellent tool for gathering user data, it is not very appealing to them. Nobody is excited to respond to questionnaires, right?
So apart from efficient questions, it is essential to consider how to increase the chances of users responding to your survey.
- Duration: Ideally, don't create a survey that takes longer than 8 minutes to answer;
- Structure: the order of questions also influence participants' engagement. Try starting with general questions, add complexity in the middle and finish with lighter questions again;
- Rewards: offering rewards can be a great strategy to recruit participants. However, be careful because it can also generate bias; people expecting a reward might keep their feedback positive only.
4) Analyze the results
Analyzing results is when the survey's effectiveness will be validated or not.
Here's when you'll see if the survey gathered sufficient data so the team can respond to the research's initial problem.
For qualitative answers, consider:
- read answers more than once to be sure of what they meant to say;
- organize data into clusters;
- analyze the results crossing them with the clusters.
For quantitative analysis, try to:
- quantify the answers, working with probability and statistics;
- compare the results with the initial hypotheses or with the results of past research;
- identify opportunities and use data to support arguments and decisions.
When analyzing quantitative data and drawing comparisons, it might be appropriate to consider Statistical Significance.
This concept refers to the probability that the results have not occurred by coincidence.
Usually, randomness is represented by the letter "p" and must be less than 5%.
Some researchers demand this value to be less than 1% depending on the product—for medical studies, for example.
In other words, statistical significance considers that a part of the survey's results is a coincidence, a random chance.
The point is that researchers have to ensure that this randomness is so minimal that it does not interfere with the results and further decisions.
UX Design and iteration are inseparable!
Iteration is a process that considers continuous development and improvement, and we can find it in methodologies like Agile, Design Sprint, and Lean UX.
Running initial research is crucial to understanding the current situation of your product, but running further research helps monitor the decisions taken to improve the user experience and the product's usability.
Therefore, iteration in the survey process is essential to keep the design team up-to-date with the product's performance.
Last tips for conducting a UX Survey
From a broader perspective, creating and conducting a UX Survey doesn't seem too complicated. All you need is a problem, questions, and recruiting people to respond.
But, putting UX Survey into practice can be tricky.
So, here comes a list with a few tips and good practices for you to consider when conducting UX Surveys.
1) Keep it simple
Remember that surveys are not supposed to be tiring or long enough to scare people away. In this sense, think of research as a product that must provide a good user experience for users.
- writing questions simply and understandably;
- making necessary questions only;
- avoid providing unnecessary information to participants;
- creating short questions.
2) Test the survey
Before surveying users, you should run a little test on a smaller group. This way, you can evaluate the efficiency of the questions.
However, testing can be easily neglected when the time is short.
Important: tests will raise your success rate, so contrary to what one might think, not running one won't buy you extra time.
3) Keeping track of costs and your budget
Before starting a UX Survey is good to know how much is the project budget for the survey and the possible costs involved.
Honestly, having all the perfect conditions to run a survey is rare. It depends on your schedule, the company's maturity, and the context of your product, among other factors.
So, remember that the commonplace in any project is making hard decisions, which is also valid for the UX Survey process.
4) Watch for biases
We have mentioned that creating survey questions is more complex than it seems.
It would help to watch for biases because they can influence users' answers.
In this sense, be aware of the following:
- Confirmation: when the question confirms a hypothesis;
- Framing effect: the way how you present a question influences its responses;
- Priming: when the first questions affect the following ones;
- No responses: when the profile of people who didn't respond is very different from those who did;
- Position effect: people tend to take the first questions more seriously than the following ones;
- Clustering: people see patterns where there are not.
These are only some biases that can affect how you write your questions. In time, study and practice will help you to be more neutral when creating questionnaires.
UX Survey and ethics
Working with people's data and information requires a level of trust and, most of all, ethics.
In this regard, when running a UX Survey look for:
- explaining the survey's objective and purpose for respondents;
- telling the company's name, or at least the market it works on;
- describing how the collected data will be used;
- knowing if the respondents feel offended by any of the questions;
- having empathy with the respondents;
- not judging any respondent's data or information.
Ethics comes first, no matter the project objectives, schedule, and stakeholders. Therefore, allow ethics to guide your actions.
We hope this article can help you prepare and conduct an efficient UX Survey for your project.
In addition, look for studying different approaches and tools to improve your knowledge of the subject.