The Ultimate Guide to UX Research
A website is the vehicle through which your client establishes brand visibility, boosts credibility, educates the public on their offering, and generates income. That said, are your clients the ones you’re actually designing websites for?
Not really. A website represents and serves the brand, but it’s ultimately the end user who engages with the website that you’re designing for. The tricky thing about website design is how do you create a website for someone you’ve never met. This is why UX research plays such a big part in the UX design process—it gives you a direct line to the user, even if you haven’t started work on the website.
In this guide to UX research, we’ll explain what it is, why it matters, as well as the various types, methods, steps and tools involved in it.
What is UX research?
Being able to create an aesthetically pleasing website is important. That’s why there’s an entire discipline dedicated to UI design. However, there’s more to the UX design than what users see. It needs to feel natural and be intuitive enough that they can easily achieve their goals.
User experience research is the practice of collecting data on users, studying their interactions with products and designing experiences based on that user input. Essentially, the goal of UX research is to help web designers create better products, like websites and mobile apps.
Why is UX research important?
UX research is an integral part of human-centered design. It enables web designers to base their decisions on facts and observations instead of on assumptions or personal preferences.
Here are some of the positive outcomes you can expect when you make UX research part of your design process:
1. Improve the website user experience
The brands you build websites for know their target users well enough to build products or services that they want or need. But, that doesn’t mean they understand their users as it relates to websites.
UX research allows you to go directly to the source and learn about their behaviors, thoughts, and goals, providing insights about your hypotheses you won’t be able to get anywhere else. This, in turn, will enable you to improve the user experience for the target user.
2. Take a data-backed approach to fixing issues
UX research is part of the lengthy and iterative UX design process taking place across many phases throughout a single project.
Since one of the goals of UX research is to create more satisfying experiences (that lead to positive outcomes, like conversions) you can continue using UX research methods to discover new problems after the start of a project. Once you have prototypes or a live website to share its a new opportunity to learn about users’ pains, motivations and behaviors.
3. Create better websites by today’s standards
The web has undergone massive changes over the years. Some of these changes have been in response to evolving technologies, other times, web design trends change because tastes require a different approach. Inclusive design, for instance, is in response to a growing social awareness and need for accessibility. Humane design, on the other hand, has arisen to mitigate the negative side effects of users spending so much time using their devices.
What connects today’s design trends and techniques is the empathetic component. Design is responding to the human condition, and UX research is the process that allows this to be possible.
4. UX research helps everyone involved
Besides just UI and UX designers, other teams on a single project can benefit from UX research. For instance, what you uncover in your data gathering and observations may be useful for:
Customer service representatives
Product designers or service providers
As a result, UX research can have far-reaching effects on the experiences that consumers have with brands and, consequently, improve brand affinity.
5. Feel confident that what you’re designing will improve business
Better user experiences are better for business, too. With research and insights applied to your website design, for example, you can feel confident in how users will interact with it. By creating an end-to-end experience that’s custom-tailored to visitors, they’re more likely to become long-standing customers and even advocates of the brand.
Types of UX research
There are different ways to classify the various methods of UX research, and the best UX research approach often employs a mix of methods. Certain kinds of research are beneficial to different phases of UX design, so having a solid understanding of what they are will help you devise your strategy.
Qualitative vs. quantitative
Quantitative research helps you gather data on what is happening. You can measure these insights with numbers and percentages. Qualitative research, on the other hand, helps you gather data that explains why it’s happening—insights that can be measured with observations and personal responses.
Ideally, you’ll use a combination of both methods. This will allow you to see the bigger picture of how your target users respond to your website design and why.
Behavioral vs. attitudinal
Behavioral research allows you to observe how users interact with the experience you’ve designed for them. This kind of research should be used to see what happens when you put users in various scenarios.
Attitudinal research allows you to determine what the users’ feelings or thoughts are about an experience. With input regarding their opinions, beliefs, and feelings, you’ll understand their general impression of your website.
These methods should be used in conjunction with one another. Behavioral research will shed light on what users do and then attitudinal research will help you understand why.
Generative vs. evaluative
Generative research helps you gain a deeper understanding of users by identifying their pains and defining their most common problems. This kind of research is useful before you design a website, since you can anticipate these issues off the bat.
Evaluative research is used in later design phases, as well as in post-launch and when redesigning websites. These kinds of research methods help you evaluate an existing website or feature and form a working hypothesis regarding how to improve it.
UX research methods
There are many different UX research methods, so don’t think of the following as a checklist of what you must do. UX research will differ depending on the specific project and objectives you’re working on.
Surveys give UX researchers the chance to learn more about their users as a whole. The questions posed to respondents can be qualitative in nature (e.g. “What does this logo make you think of?”) or quantitative (e.g. “How would you rate the ease of use of this contact form on a scale of 1 to 10?”).
The nice thing about surveys is that you can gather information about a group of users while expending minimal effort and time.
The one-on-one user interview is another method focused around a question-and-answer format. However, this method establishes a direct interaction between the UX researcher and the user.
There are different types of user interviews you can conduct:
Directed interviews: The researcher asks specific questions related to the product, which enables them to focus on specific components of the website or parts of the user journey. This can be used to gather both quantitative and qualitative data, and really flesh out how users feel when they see or interact with your product.
Non-directed interviews: The researcher has an open conversation with the interviewee about the product. The goal of these interviews may be to learn more about the users’ general attitudes, pains, and goals with regards to certain product features and functionality.
Focus groups are like user interviews, except that the interviewer engages with a group of users all at once. The discussions also tend to be managed by a neutral moderator instead of a researcher.
The purpose of a focus group is to gather qualitative, attitudinal data from users. The moderator conducts a structured interview to learn more about the users’ views, attitudes and experiences with relation to the product (either the one that’s being built or existing solutions).
Ethnographic studies are ones where the researcher observes—and sometimes engages with—the user in their natural environment. These kinds of studies don’t need to be done in-person. Researchers may remotely observe the user (for example, over Zoom) as they interact with a product.
The purpose of these studies is to gather behavioral data and figure out how users will naturally engage with a website. This kind of observational data can’t be used on its own though. It needs to be gathered in conjunction with attitudinal data to ensure that how users feel or think about a website translates to how they engage with it, and vice versa.
This is another type of observational study. However, with diary studies, observation takes place after the fact. In this case, UX researchers give target users a physical or digital diary to self-report on their activities as it relates to the website and how they felt along the way.
In order to gather meaningful qualitative data from a diary study, it must take place over a long period. This gives the researcher the ability to identify trends in the users’ behavior and attitudes.
Card sorting is a qualitative research technique that’s useful in creating the information architecture of a website. UX researchers set up either physical or digital cards that represent the various content on the site, and ask users to provide feedback on them.
There are different types of card sorting. The method you choose depends on which parts of your information architecture you need help with figuring out. For example:
Open card sort: Users are given a set of labeled cards. They’re asked to group them in the way that makes the most sense and to give a name to each category.
Closed card sort: Users are given both a list of categories and a set of labeled cards. They’re asked to organize them under the given categories.
Hybrid card sort: Users are given both categories and labeled cards. While they are asked to organize them using the categories, they also have the option to create their own.
Tree testing is another research method used to sort out the website’s information architecture. Specifically, you’ll use these tests to optimize your sitemap. In tree testing, users are given the top-level of a sitemap, and the researcher asks them how they would carry out a task.
If they’re given a sitemap prototype, users are able to click on the links to demonstrate the steps they’d be inclined to take in real life. If a prototype isn’t yet ready, they’ll describe their thought process in approaching it. Either way, these insights will help the researcher improve the way they label and organize the navigation of their site.
Usability tests are similar to tree tests, except they go beyond the sitemap and take a look at how users think about and engage with a website and its features. Researchers can use usability tests during the initial research phase to learn how users prefer having key components designed. They can also perform usability testing after a prototype has been developed or launched to diagnose issues with the UI or UX.
Here are a couple types of usability testing you can do:
Moderated usability tests: Can be done in person or via screen share. The UX researcher asks the user to complete a task. The user then explains how they’d do that or demonstrate using a prototype. They’re also encouraged to explain why they’re doing what they’re doing.
Unmoderated usability tests: Use session recording software. The user is provided with instructions on what to do and which tasks to carry out. Again, they’ll go through the motions while explaining their thought process. The researcher reviews the results later.
A/B testing is a type of quantitative UX research method that takes place after a website has gone live. A/B testing software is installed on the website, which enables the web designer to create an alternative design (or designs). The different versions are shown to an equal number of website visitors.
A/B testing can be used to:
Help the researcher decide between competing designs or functionality
To remove friction from the website and replace it with a more user-friendly design
To learn how to optimize the website for greater conversions
As a general rule of thumb, A/B tests are done on one specific segment of the website, like a contact form’s fields, a button design, or a hero image. While UX researchers and designers tend to do lots of A/B testing to improve a website, only one test can be run at a time.
When used in conjunction with A/B testing, heatmapping can be a useful tool in website optimization. Heatmapping is both a qualitative and quantitative UX research method that takes place on a live website.
In heatmapping, software monitors how users engage with a web page by visualizing the following data:
How “hot” (heavily viewed) specific areas of a web page are
How far down a web page the majority of visitors scroll to
Which areas of the page people click on most and least
Researchers and designers use this visualized data along with engagement percentages to make inferences about the success of their content and the user journey.
5 steps to conducting UX research
User experience research isn’t all that different from the scientific method. While UX researchers and designers use methods that are specific to websites, the general process they follow is similar to scientific researchers:
Step 1: Choose the objective
User experience research can be done at any step of the design process, even after the launch of a website. As such, there are many different goals you may be looking to achieve through data-gathering methods and analysis. Ultimately, what you need to ask yourself in this step is:
What is your main objective in doing this research?
What specifically do you want to find out about your target users?
For instance, let’s say you’re considering adding a chatbot to your website. You might ask: Will visitors user this chatbox? What is the best place to put it on my site? And what chatbox design or content will best engage with and help viewers?
Step 2: Formulate a hypothesis
Once you have a goal, you’ll need to structure it a bit more. Take, for instance, the chatbot example. There are many ways to go about exploring this feature, so you’ll have to zero in on one specific part of the experience, like how many visitors engage with it or how users feel about it being on the website in general.
This process will first require you to examine what you already know. Then, create a hypothesis about what you think you’ll observe in your research or what you’ll measure in your tests.
Here’s how you might flesh out the chatbot hypothesis: Chatbots are a common feature on eCommerce sites because they allow shoppers to connect to live customer service representatives using a chat window. When we ask users “How would you contact customer support from this website’s home page?” the majority of them will answer that they click on the chatbot.
Step 3: Determine the best UX research methods to use
It’s important to understand the different types of research methods there are so you can narrow down the list of methods to use based on the data you’re seeking. They can be narrowed down further based on the kinds of resources you have available to do the research and the means to carry them out.
It’s absolutely a good practice to use multiple methods. This will ensure you measure the data from different angles—like a quantitative method and a qualitative one.
Using the chatbot/customer support hypothesis, for example, you could use the following methods:
Survey to identify the top three methods among target users for contacting customer support on a website.
Ethnographic study to observe and get a sense for why users complete the requested action the way they do.
Step 4: Use the right tools to research phase and collect your data
While you as the researcher or designer have to develop the experiment (and sometimes actively run it), UX research isn’t a completely manual practice. There are tons of UX research tools that can help you set up and run tests and, later, analyze the results.
In the next section, we’ll share what some of those tools are. Use the ones that make the most sense in helping you conduct your research.
Step 5: Analyze the results and implement a plan
Once you’ve done the research and gathered your data, you’ll analyze the results within the context or your original goal. Ask yourself the following questions:
Did you complete the objective in full?
Was your hypothesis correct?
If the research didn’t generate relevant data or provide you with new insights, was it due to the types of UX research methods pursued? The participants recruited? Something else?
What will you do now that you have this data?
Even if the results didn’t turn out how you hoped they would, there’s always something to learn. Take those lessons and use them to improve your process, your website, and your user experiences.
UX research tools
User experience research can be a time-consuming and costly process. However, that shouldn’t keep you from doing it.
One way to decrease the time and monetary costs of conducting UX research is to leverage the UX and UI design tools that help you recruit users, run sessions, and synthesize data. Here are eight of the best multipurpose UX research tools and the types of user research methods they’ll help you with:
dscout: participant recruitment, user interviews, diary studies
Hotjar: surveys, heatmaps, website session recording, user feedback
Optimal Workshop: participant recruitment, card sorting, tree testing, usability testing, surveys
SurveyMonkey: surveys, questionnaires, user feedback
UsabilityHub: usability tests, click tests, tree tests, surveys
Userlytics: participant recruitment, user interviews, focus groups, card sorting, tree testing, usability testing
UserZoom: participant recruitments, surveys, user interviews, usability tests, card sorting, tree testing, click testing
UXtweak: participant and focus group recruitment, prototype testing, website testing, surveys, session recordings, card sorting, tree testing
By Jenna Romano
Web Design & UX Expert, Writer