Google UX Design

Explore

UX Research

Planning UX Research studies

Understand UX Research Process

A research study is a step-by-step examination of a group of users and their needs, which adds realistic context to the design process.

UX researchers adopt various methods to uncover problems and design opportunities. In doing so, they reveal valuable information, which can be fed into the design process.

To get started, step 1 is plan the study, step 2 conduct the research, step 3 analyze and synthesize the results, and step 4, share and promote the insights.

To plan a study, start by outlining the background for the project.

Next, set goals for your research, and write down the questions you want to answer. Then establish the steps you'll take to conduct the study and select the people who will participate in the study.

The second step is to conduct the research. During this step, you'll gather data. There are several techniques you can use, but the one we'll focus on in this course is called a usability study.

A usability study is a research method that assesses how easy it is for users to complete core tasks in a design. The goal of usability study is to identify pain points that the user experiences with your designs so the issues can be fixed before the final product launches.

During a usability study, you get a chance to see how users interact with your new product or feature. You can also interview users to learn more about their experience.

The third step is to analyze and synthesize your results. This involves trying to find the actual meaning in the data. You want to figure out why the data is the way it is. Look for patterns in the quantitative data and explore trends in the qualitative data from participants' answers to interview questions.

The last step is to share and promote your insights with the project's stakeholders. Project stakeholders are people who are involved in the project or who will be impacted by its results. Project stakeholders need the results of your research and need to agree with the direction of the project.

⁠

Using language that’s appropriate for the context can help you get a deeper understanding of participants’ unique perspectives and values. Every word you use when asking questions, taking notes, or transcribing quotes needs to accurately represent the ideas that participants convey.

Your understanding of the best language to use in any given situation will continually shift as you learn the ins and outs of conducting UX research. Using appropriate language while conducting research is an ongoing process!

There’s no right or wrong way to communicate an idea, but language often reveals hidden power imbalances. When conducting research, you want to make sure you’re communicating in a way that’s inviting, not commanding. For example, avoid using words that are considered ableist, or words that assume a person does not have a disability. You can imagine that a researcher might say to a participant, “Let’s walk through the details of the study.” Instead, it’s just as easy to say, “Let’s go through the details of the study.” That second phrase doesn’t make any assumptions, intended or not, about a person’s physical capabilities.

Build a UX research plan

⁠

There are seven elements that your plan should include: the project background, the research goals, the detailed research questions, the key performance indicators or KPIs, the methodology, the participants, and the script or questions you'll ask participants.

The first element of your plan is the project background.

The project background answers the question: What led you to conduct this research?

It's a brief explanation you can give to anyone who asks why you're doing this research.

Identify the signals that indicated research was necessary.

Ask yourself why you're doing this research and if there's a problem you're trying to solve.

Describe any previous research that has been conducted or solutions that have been tried.

How have previous attempts brought us closer to solving the problem?

Lastly, list insights the research will generate.

An insight is an observation about people that helps you understand the user or their needs from a new perspective.

A great insight inspires clear action and uses simple language.

In the project background, include how the insights will be used and what decisions will be made based on those insights.

“We're creating a new app to help people find and schedule dog walkers.

We need to find out if the main user experience, finding and scheduling a dog walker, is easy for users to complete.”

Next, your plan should include research goals.

Ask yourself questions like, What design problems are you trying to solve? And how will the results of the research impact our design decisions?

Research goals state the specific ideas that you want to learn from the research or what you would

like the outcomes of the research to be.

Use your answers to create goals for your research project.

“Determine if the dog walker app is difficult to use.”

Research at this time is called foundational research.

The goals of the research are to better understand the user problem you're trying to solve.

You want to make sure that there's a real need for the product.

We need to consider the ease of use in order to understand why customers join our app and leave or stay.

After you determine research goals, you need to develop detailed research questions for your plan.

What are the questions your research is trying to answer?

“How long does it take a user to find and book a dog walker in the app?

And what can we learn from the user flow or the steps that users take to book a dog walker?”

Finally, if you run research on a product after it has been launched, your research goals help you understand if the product worked as expected.

This type of research is called post-launch research, and the goals reflect whether you have successfully completed what you set out to do.

Instead, the research questions should be the questions you want your research to

answer and should align with the goals of your research.

Why are research questions important? They guide your research.

Research questions provide focus and structure for your research study.

A good research question should be actionable. You should be able to identify a clear way

to attempt to answer the question. You want to answer specific questions and produce

meaningful data.

Questions should be neutrally phrased so that they don't sound like you're assuming a particular answer to your question.

“On average, how many times a week does a user hire a dog walker through the app?”

“What frustrated users most about finding a dog walker through the app?”

Quantitative research focuses on data that can be gathered by counting or measuring -- think numbers.

Qualitative research focuses on observations about why and how things happen -- think written descriptions.

Another important element of your plan is the key performance indicators, or KPIs.

As a reminder, key performance indicators are critical measures of progress toward an end goal.

You might ask, how can you measure your progress toward the research goal?

For our app, one thing we should track is how many users in the research study complete their search for a dog walker.

So the KPI would be the percentage of users who book a dog walker.

The next element of your plan is the methodology. This is where you document the steps you'll take to conduct your research.

How will you collect data and how will you analyze the data once you get it? To find out if users are able to find and schedule a dog walker in our new app, we might want to conduct a survey of prospective users. To analyze the survey results, we will use a spreadsheet and identify trends.

In addition, your research plan lists the research participants.

Who will you survey? What characteristics do the participants have?

For example, you might want to recruit participants who are dog owners with full-time jobs and who go out for activities more than once a week. The types of participants you select should be based on your research goals. Also, make sure that participants you select do not bias your results.

When conducting research, we need to choose study participants who are representative of all our users, not just a select group.

⁠

Key performance indicators

“How did the research study you conducted last week go?”

Any time you conduct research, you want to have a way to measure the effectiveness of your product or prototype.

To do so, use key performance indicators (KPIs), which are critical measures of progress toward an end goal. As you might remember, KPIs are the fourth element in a research plan, and you’ll often be asked about your KPIs as a UX designer in the real world.

⁠

Time on task measures how long it takes for a user to complete a task. A task can involve any function within the product you’re designing, such as filling out a form or making a purchase.

To measure time on task, all you need is a timer! Start timing the user when they begin the task you’ve assigned, and stop timing as soon as they completed the task. For example, you might time how long it takes for a user to start from the home screen of your app and complete the checkout flow to purchase a shirt. It’s generally safe to assume that the less time it takes for users to perform a task, the more effective your design is.

⁠

Use of navigation vs. search indicates the number of people who use a website or an app’s navigation, compared to the number of people who use the search functionality. In other words, some users will prefer to use the navigation bar to get around your product, while others will go straight to the search bar, type in a query, and be directed to part of the product.

To measure use of navigation vs. search, count the mouse clicks or taps on navigation-related parts of your design, and compare that to the number of times a query is entered into the search bar. This KPI, unlike most of the others, measures the preference of users, instead of whether something is “good” or “bad,” so you don’t need to worry if the numbers are high or low. Instead, the goal is to help you understand how users interact with the navigation and search functions of your product. Keeping track of your users’ preferences can help you strike a good balance between them in your designs.

⁠

User error rates indicate the parts of a design that cause users to make errors. For example, a user might click on the wrong icon when trying to make a purchase, forget to check a box, or submit incorrect information when filling out a form. These errors are not the user’s fault! Instead, user error rates help point to areas where you need to make improvements to the design of the user experience.

To measure user error rates during a research study, keep track of the parts of your design where users make mistakes when completing the tasks you’ve assigned. As a general rule, the lower the number of errors, the better the design.

⁠

Drop-off rates show how many users abandon the experience. In other words, this KPI reveals how many users quit before reaching the end of a purchase or some other endpoint you’re trying to lead them to. Users might quit using your product if the navigation is difficult to understand, if they get frustrated while trying to complete a task, or if they simply get bored.

To measure drop-off rates in your own designs, count the number of participants who quit a task or don’t make it to the end of their goal. Then, make changes to your designs to improve the user experience and run a second research study. You can compare the drop-off rates from each study to measure the success of the design changes. Your goal is to decrease drop-off rates with each design iteration.

⁠

Conversion rates measure the percentage of users who complete a desired action. Conversion rates are the opposite of drop-off rates. Any time a user successfully completes a task, meets a goal, or makes it to the final destination of your product, it’s a conversion. Think of the dog walker app as an example: The user needs to take multiple steps to find and reserve a dog walker. The conversion rate will show the percentage of users that actually made it to the end of the flow and booked a reservation.

To measure the conversion rate for your product, count the number of research study participants who complete a listed action. As a general rule, the higher the conversion rate, the better the design. In addition, comparing the conversion rate from one research study to the next can help you gauge the success of the changes you made to your designs.

⁠

A System Usability Scale (SUS) is a questionnaire that asks participants their opinions about your product; the results are used to measure the usability of your designs. In an SUS, users are asked the extent to which they agree or disagree with 10 statements about the usability of a design. For example, users might be asked to respond to the statement: “I thought the app was easy to use” on a scale from “strongly disagree” to “strongly agree.” It’s a quick and reliable way to know if a design is working.

You can use a SUS to measure the usability of your own designs! You want participants to “strongly agree” with positive statements (such as “I thought the app was easy to use”) and “strongly disagree” with the negative statements (such as “I found the design unnecessarily complex”). You can also calculate an overall score for your design based on the survey responses.

⁠

The Net Promoter Score (NPS) is a measure of how loyal users will be to your product or service. The NPS measures the likelihood that a user would recommend your product to a friend or colleague. Participants rate the question, “Would you recommend this product to a friend or colleague?” on a scale of 0 to 10.

Promoters are participants who give a rating of 9 or 10, which means they would recommend your product to others.

Passives are participants who give a rating of 7 or 8, which means they are satisfied with your product, but they probably won’t pass it along to friends or colleagues.

Detractors are participants who rate from 0 to 6, which means they could warn people away from your product or service.

To calculate the NPS, subtract the percentage of detractors from the percentage of promoters. When your NPS is positive, it’s a sign that users are satisfied with your designs. Well done! You want your NPS to be a high positive number. If your NPS is negative, it’s a sign that your user experience might have bugs, be difficult to navigate, or might otherwise cause frustration among users.

Choosing KPIs

These seven KPIs are important measures of success for your research study and for the designs you're testing. When deciding which KPIs to measure for your project, think about the goals of your research and the findings you want to be able to present to your team, or include in your portfolio, after the research concludes. Each KPI is unique, and together, they work to help you measure the success of the user experience you’ve designed.

⁠

Methodology

Methodology involves the steps you’ll take to conduct research, collect data, and analyze data. When thinking about the methodology, it’s important to consider 1) the type of research method involved, 2) the steps involved in the research method, and 3) the type of data that will be generated from the research method.

Comparing research methods

Primary research is research you conduct yourself, while secondary research is research that uses information someone else has put together. Some examples of primary research are interviews—which you conducted in Course 2—surveys,

competitive audits⁠

, and usability studies. Another point to consider is whether the research method is quantitative or qualitative.

Qualitative research focuses observations on why and how things happen, and quantitative research focuses on data that can be gathered by counting or measuring.

Conducting research with a usability study

This research method is ideal at this stage of a project because it involves testing out your design with real, potential users.

Usability studies are primary, qualitative research methods because they allow UX designers to make direct and indirect observations based on the participants’ behavior and words. You’ll give each participant a few guided tasks and take notes on how successfully the tasks are completed (direct observation), and you’ll also record the study in order to play it back later (indirect observation)to discover new insights from the study.

Participants

Recruit a diverse participant pool

The participants in your study should represent your product’s potential users. These participants play a big role in shaping your designs because you’ll take their pain points and suggestions into account as you iterate.

Where can I find participants?

If you’re conducting research and creating designs for a company with an existing user base, you’ll likely be able to recruit participants from that group of established connections.

Alternatively, if you’re coming up with designs for an imaginary company or a newly established business, the easiest way to recruit participants for your study is online. You can use social media, online design communities, or even fee-based research recruitment agencies. There are also websites created specifically to connect companies with research participants, like

UserTesting⁠

and

User Interviews⁠

. Keep in mind that recruiting participants online can have downsides. For example, you risk leaving out populations of potential users, like older people who don’t use the internet or people who don’t have internet access at home. As you create your research study plan, consider the target audience for the product you’re designing and decide if you can recruit that type of user online.

If recruiting participants online isn't an option, a less formal way to recruit for your study is to ask people in person. You can use a recruiting method known as hallway testing, which means asking people that pass by in the “hallway” to try the product you’ve designed.

Incentives are something that motivates or encourages people to participate in a research study. Incentives can take the form of monetary compensation, gift cards, physical gifts, or a gesture to thank participants for their time and feedback, like a paid lunch. No matter where you find participants — online or in person — you’ll need to consider how you’ll incentivize them to engage in your study.

The best way to check if potential participants meet your desired characteristics and represent a diverse set of individuals is with a screener survey. A screener survey is a detailed list of questions that help you determine whether potential participants meet the requirements of the research study. To learn more about screener surveys, check out this

article from User Interviews⁠

A representative sample is a small group of participants who represent both your target user group and user groups that are often marginalized. And a marginalized population is one where people have specific characteristics and life experiences that prevent them from fitting into what society inaccurately defines as “normal,” such as people with disabilities or people with limited access to technology.

Instead of recruiting participants who are visually impaired, hearing impaired, or who have physical impairments, recruit participants who use the assistive technologies you want to involve in your study. For example, seek out participants who use:

Screen readers

Closed captions

Switch devices

Keyboard-only navigation

Magnification devices or magnified screens

Any other assistive technologies that will provide insights about your UX design

⁠

Assistive technology for participants with disabilities

As a reminder, accessibility means the design of products, devices, services, or environments for people with disabilities. One of your key priorities as a UX designer is to create products that are usable and accessible to all people.

Sometimes, the abbreviation a11y (pronounced “A eleven Y”) is used to refer to accessibility. It’s shorthand for representing the 11 letters between the letter “a” and the letter “y” in the word “accessibility.” More importantly, a11y resembles the word “ally,” which reflects the idea that UX designers need to be allies to people with all types of abilities.

It’s critical that you consider and involve people with various accessibility needs when designing, especially during the empathize and test phases of the design process.

⁠

A permanent disability is one that affects a person long-term, like losing sight, hearing, speech, or mobility. For example, Amir, a person with permanent blindness, uses a walking stick to navigate their surroundings.

A temporary impairment is a short-term illness or ailment that can be caused by an injury or other limitation. Consider Margo, who has temporarily blurred vision without their glasses.

A situational challenge occurs when a person’s environment blocks certain functions. For example, Juan cannot read text messages on their mobile phone while driving a car. Instead, Juan uses voice commands to hear and send texts while driving.

⁠

An assistive technology (AT) is a product, equipment, or a system that enhances learning, working, and living for people with disabilities. In the examples above, assistive technologies came in the form of a walking stick, glasses, and voice commands. For your research study, you should strive to include as many different users of assistive technology as you can.

Let’s explore some of the most common ATs.

A screen reader is an AT that interprets and verbalizes text, button names, keyboard strokes, and code that a website or app is composed of. Screen readers are often used by people with low vision. In addition to screen readers, some people with low vision might use a computer or smart device with a high contrast screen or increased magnification.

A switch is an AT that helps people with disabilities use technology - like computers, phones, appliances, and mobility equipment - with minimal movements and gestures. A switch can come in many forms, like a button or clicker. For a computer, a switch might replace a traditional keyboard and mouse.

Closed captioning and speech-to-text are both ATs that convert audio into text for people with limited hearing.

Reminder alarms with simplified text and supporting images can help people with cognitive disabilities remember important information. For example, Android phones have a feature called

Action Blocks⁠

where users add common actions to their home screen with a name or image. So a photo of your mother on the home screen of your phone will call her phone number.

Augmentative and Alternative Communication (AAC) devices are ATs that support people with cognitive disabilities who may experience speech limitations or learning disabilities by using images to communicate instead of words.

Script

A script, also called a discussion guide, is a list of tasks and follow-up questions a researcher presents to participants in the interview portion of a usability study. Using a script in your research plan gives you a clear understanding of what a user is trying to do, how they think and feel, and what their problems are. Making a habit of writing scripts for your research plans will improve your ability to understand your users and identify solutions that address their needs. The tasks and follow-up questions in the script should be based on the research questions and KPIs you chose earlier in the study plan.

A well-rounded script is:

Open-ended: Task and follow-up questions encourage discussion and reflection rather than short "yes or no" responses

Objective: Tasks and follow-up questions don’t try to lead participants toward a particular answer or introduce the researcher’s personal bias

Goal oriented: Tasks and follow-up questions relate directly to the research goal

Want to print your doc?
This is not the way.

Try clicking the ⋯ next to your doc name or using a keyboard shortcut (

CtrlP

) instead.