Understanding UX Research in 20 minutes
An outline of one of the most important aspects of UX Design
Properly carrying out UX design research is critical to the success of any design project. It will help you define your problem, validate your assumptions and ultimately produce a sweet design.
Sounds like a good deal, eh?
But if you have no idea where to start, research can be a daunting task. Like the first time you popped a pimple — it needed to be done but you just didn’t know where to apply your efforts.
Luckily, adhering to best practices of design research is not rocket science. Or brain surgery. There are some general guidelines and principles that all research projects should incorporate. Once you start doing research, make mistakes and gain experience, you’ll not only internalize the process of design research, but you’ll also master its art form.
Whether you are just setting sail on a career in UX design or a seasoned vet, I hope this article will outline the critical aspects of great design research at a high level so you can get started right away.
And here’s the disclaimer that all articles of this nature must have: this post is by no means exhaustive, should not be taken as gospel and will imminently evolve. Please refer to the end of this article for additional resources and perspectives on design research.
One more thing: I want to give proper credit to the inspiration for this post, Erika Hall’s Just Enough Research. To fill in the gaps of what this article misses, please go purchase her book. It is worth it. And it’s short.
Ok, cool. Let’s get started.
What is research, anyways?
Research is simply focused and repeated questioning that aims to answer questions surrounding a specific goal.
When you design something, you always have a goal in mind. Let’s say, for example, you want to effectively sell man bun related products for a website called “Mr. Man Bun.” But when you start out, you’ll have a blank slate. You won’t know how to accomplish that goal (even if it’s as worthy as selling man bun products).
Research is an irreplaceable component of the design process that answers your burning questions and sets you on a more enlightened path towards a great design.
Research is a process, which means there are certain things that you want to be doing each time. Hopefully this article can serve as a reminder of those things you gotta do. The process will aid in asking better questions, in a better way, to an appropriate audience. It will help you analyze your findings and turn them into actionable insights.
Yet, research is also an art. The more you do it, the better you will become. You need to find your own style and develop your own instincts. More research does not always mean better insights — know when to stop and when to keep going. Further, you need to become a master at involving a variety different people in the research process, many of whom may be completely uninterested.
But, what makes my research “good”? Glad you asked.
Throughout your process, keep in mind that good design research is…
- High quality, repeated questioning. You and your team are insatiable question-askers of thought-provoking questions. This isn’t the traditional doctor’s office interview, where you’re asking standard questions and getting standard answers. Great insights will come from asking tough questions that make your users think. Further, as we will touch on later, consider asking questions within the context or environment of your end users.
- Collaborative. Just because you have one researcher on your team doesn’t mean one person is doing the research. Each team member will have a unique perspective and thus a unique question. A developer might be interested in a user’s level of tech experience, and a graphic designer might probe the user’s aesthetic tastes. Drawing on multiple perspectives will undoubtedly make your design more robust.
- Seeks understanding of your users. The crux of good design research is to understand the motivations and goals of your end users. Be empathetic and flatter them with questions. People love to talk about themselves — give them a reason to spill their frustrations, ambitions and secrets. But don’t be nosy. Or creepy.
- Well defined and prioritized. First of all, list the questions, goals and assumptions that your design will address. It is important to understand what questions you need answered, and where you are making informed guesses. Prioritize these. Which questions, goals and assumptions are absolutely crucial to the success of your design? For example, figuring out whether men with man-buns actually care about man-bun products might be high up on your list of things to validate.
- Open-minded, skeptical and empathetic. You must empathize with your end users. To use the overused but useful cliché: put yourself in their shoes. Observe and interact with your end user in their natural habitat. Context is key. You must also keep an open mind throughout the research process. We humans are inclined to draw conclusions from incomplete data. Try to avoid this by refusing to articulate your conclusions until you’ve collected all the data you’ve intended to collect. Always poke holes in your own assumptions and conclusions to see if they’re truly solid or merely flimsy guesses.
Write these principles down and put them on sticky notes somewhere where your team will see them. Since some of this stuff isn’t entirely intuitive, you’ll need some reminders of how to stay the course. With these principles in place, it’s time to get dig into the meat of design research.
Here are the stages we will go through:
- Define your problem(s) and questions
- Plan how you will get answers
- Do your research
- Analyze your research
- Produce a dope design
- Distribute hi-fives
Defining your research
The absolute first thing you must do when conducting UX research is to clearly state a set of goals or questions that you need answered. From our example with Mr. Man Bun, I might want to explore the following:
Why do men choose to grow out their hair? Why man buns as opposed to another hairstyle?
How do men manage their man buns?
Analyze the relative advantages and disadvantages of man bun product competitors.
It is important that:
- Your goals are specific. Being vague and unclear is the enemy. For example, saying something like, “We will explore the world of man bun products,” is listless and ambiguous. In contrast, “We will analyze the relative advantages and disadvantages of man bun product competitors,” is specific and actionable.
- Your goals support the business. At the end of the day, your design work is useless if it doesn’t help your client make money. Finding out something like, “How do men manage their man buns?” is an opportunity to identify products and services that men with man buns need. Or don’t need.
- You define a set of users. We will delve more into this in a minute, but it is important that you identify your users right off the bat. You’ll need someone to answer your questions, after all. For Mr. Man Bun, we will be talking to men with manly buns of hair.
Now that you have your goals and questions defined, it is time to get more specific on who will be answering these questions. To accomplish this, it is crucial to understand the concept of a stakeholder. A stakeholder is anyone or anything that has some sort of relationship with the business and goal(s) you have in mind.
Examples of stakeholders are…
- The end users. The group who will ultimately buy, subscribe or complete any other goal that the particular business is interested in. For our example, these are men with man buns who would be purchasing man bun products or services.
- The competition. Any organization or business out there that engages in selling or producing content on the subject matter at hand. This could be anything from prominent man bun blogs or men’s conditioner brands.
- Members of the business you are working for. You’ll want to understand the people for whom you are doing the research. Interviews with executives, sales support reps and marketing folks at Mr. Man Bun would be a good start.
These will be the people who will ultimately inform your design. The biggest failure of the design process is designing for just one person: yourself. The greatest success is designing for those who you have sought to understand. Some random dude named Albert Einstein once said, “A man should look for what is, and not for what he thinks should be.”
Planning your research
By now, you have a few goals in mind and you have considered your end users. Now it is time to take action and outline your research process. Just another disclaimer: there is no “right” way to do UX research. Different projects will merit different approaches. Different teams will merit different approaches. Like I said, you’ll have to master the craft. Given this fair warning, I will outline a standard method that will get you up and running.
(Insert quote here about how planning for success will bring you success)
Inspired, we move forward…
Here are the two hefty steps to planning your UX research:
1. Designate research roles among your team members.
Excellent research requires gold-medal-winning collaboration. But you can’t expect collaboration to magically happen. You must plan for it. Below are the various roles that you will need to fill. It’s ok if someone takes on a couple of these roles, as long as everyone knows what they are expected to do.
Point person & report author. The point person is the go-to research guy — the fearless leader of the researching. He will plan the research, including defining the problem (all the stuff in the previous section), writing interview scripts and providing a full written outline of the research.
Interviewer/moderator. This person is the one actually interacting with your end users. She will conduct testing and/or interview sessions.
Coordinator/scheduler/recruiter. The coordinator will be in charge of recruiting test subjects and scheduling times for them to meet with the team.
Note-taker/recorder. It is crucial for the note-taker and the interviewer to be two separate people. You want to have one person fully dedicated to responding and asking follow up questions to the user without being hampered by having to write everything down at the same time. However, if this is unavoidable, simply ask permission to record the conversation so you can go back and listen.
Observer. It is useful to have other employees from either your firm or, better yet, your client’s firm watching your interviews or usability tests. However, make sure these people are not obtrusive. If you cannot have physical observers, make your recordings available.
Analyst. The analyst role could be given to multiple people, as this is the person looking for patterns and insights in the data you’ve collected.
Documenter. This person will report the results once the study is complete.
2. Outline exactly what type of research you will do.
Once your inquisitive warriors have their assignments for the Battle of Understanding, it’s time to formulate your battle plan. In normal people prose: figure out exactly what type of research and how much of it you will do. It is important to ask your client beforehand if he has done any prior research that you can draw from. Don’t think that you have to do all of the legwork yourself — be resourceful.
Once you’ve figured out what you already have, turn back to your questions and problems and ask where the holes in understanding might lie. Let’s say our Mr. Man Bun client has extensive research on man bun product competitors, but nothing about the actual man-bun-wearers. In this case, you’ll probably want to allot more time to getting to know man-bun-wearers than collecting intelligence on competitors.
Below is a list of the types of UX research that you might conduct, separated into categories based on the type of information to be collected. For brevity’s sake, I won’t fully outline each one. So, for additional resources on each of these topics, I’ve provided helpful links in the resources section at the end of the article.
Goal: To deeply understand what your competitors do and determine where your client fits into the competitive landscape.
Competitive research is fun and inspirational. You get to pick apart someone else’s design with ruthless criticality and plunder its greatest qualities. But don’t forget the business end of things: you are here to understand where your client’s brand fits into the competitive landscape. Hint: it fits in its own, unique space.
Competitive audit. The most comprehensive way to dig deep on the competition. This includes looking at the competitor’s strategy, target market, pricing, copy, aesthetic style (colors, shapes, graphics), UX/UI, conducting a SWOT analysis (see below) and more. Here’s an example of a competitive audit I did for project management applications. Feel free to steal the format.
SWOT Analysis. If you’re going for a more agile approach, the classic SWOT analysis may be all you need. SWOT stands for Strengths, Weaknesses, Opportunities and Threats. You probably glossed over it at some point in college; now it’s time to get reacquainted. This will be a simplistic and effective way to understand your client’s position in a given market.
Competitor infiltration. I made this term up. But it sounds cool, right? In other words: be an undercover customer of your competitor’s product. If there’s a free trial, try it out. Do some usability testing on it (more on usability testing later). Make up an issue and contact their support team. Have some fun with it. You’ll learn first-hand their Achilles heel.
Goal: Uncover the habits, mental models, frustrations, desires and environments of your end users.
Note: This part is super important.
My earlier statement bears repeating: The biggest failure of the design process is designing for just one person: yourself. The greatest success is designing for those who you have sought to understand. Your design is worthless if it doesn’t make sense to your end users.
Surveys. Cheap and simple. Go to SurveyMonkey and get wild with it. Make sure the questions that you are asking help shed some light on the problem statements you outlined.
Some things to keep in mind before you release your survey to users:
- Establish trust. Introduce yourself and show your logo so that the participants know who they are completing the study for.
- Keep it simple. Surveys are annoying, so try to make it as painless as possible. Offer an incentive like a gift card if you must.
- Include a few qualifying questions that help identify who is taking your survey. You’ll only want to look at responses from people who characterize your target market and disregard the rest. For Mr. Man Bun, I’d ask the participants if they have man buns.
- Test it on yourself and your team. This will weed out any silly mistakes.
Interviews. Qualitative data is just as good as quantitative data, if done right. Take the time to learn about conducting great interviews. I won’t try to outline everything here.
However, two big things to keep in mind:
- Make the interview structured. Similar to your surveys, you want to make a list of questions that aim to unravel the mysteries in your problem statement.
- Keep it relaxed. Remember you’re talking to another human being — try to open up with some small talk. Don’t be too structured. Allow room for follow-up questions.
Contextual Inquiry/Ethnographic Research. These methods will illuminate one of the most important considerations of the UX research process: user context. Both contextual inquiry and ethnographic research involve interviewing or observing the user in his or her natural environment. People are often bad at self-reporting their feelings in interviews, so these methods will give you a chance to see your users in action. Immerse yourself in their world and try to empathize with their daily situations.
While this will be a fictional character with a made up name, you will want to base it off of someone you’ve met during the research process. User personas are incredibly important because they will give you a reference to consult while you are designing so that you can always make sure you are designing for your users, not yourself.
Goal: Get acquainted with the people, politics, processes, ambitions and attitudes of the people in the organization you are working for.
Organizational research means applying a blend of competitive and user research methods to fully understand your client’s organization. Whether you are in an in-house or agency setting, it is incredibly important that you have a full grip on the people, politics, processes, ambitions and attitudes you are working with. Design teams function best when they can play nicely with others.
Things to do:
Competitive audit. Conduct a competitive audit, or at the very least a SWOT analysis, of the company you’re working for. Get a sense of where the company has been, what the current strategy is and who the key decision makers are.
Interviews. You will want to interview not only key decision makers, but also people who might be familiar with user concerns like customer service representatives. Flatter them with questions. Understand what they want and where they see the company going. Get a sense of the politics — who they trust and who they don’t. However, tread carefully. You are trying to gather insights, not start an office catfight.
Goal: Assess the validity of your design.
Evaluative research is something you should do constantly. It means testing your design against common usability pitfalls and user expectations. Each time you create a new layout or interface, come back to this section and tick off some boxes.
Heuristic analysis. A heuristic is a rule of thumb. It’s a ready-made response to certain situations. Like when that girl Nastia follows you on Instagram and asks if you want to see her unmentionables, you calmly but decisively click back to your homepage.
The legendary usability guru Jakob Nielsen came up with the idea of heuristics for interface usability. There are ten of these bad boys:
- System status visibility. The design should give appropriate feedback about what’s going on.
- Match between system and real world. Use conventions and plain, understandable language.
- User control and freedom. Undo, redo, back, forward, home button, etc.
- Consistency and standards. If two buttons in your design look similar, they should act similar.
- Error prevention. Avoid errors in the first place, like giving smart defaults.
- Recognition rather than recall. We humans suck at memory stuff. List the options and don’t make the users remember information.
- Flexibility and efficiency of use. Shortcuts for macho users.
- Aesthetic and minimalist design. Get rid of unnecessary information.
- Help users recognize and recover from errors. Error messages should be helpful.
- Help and documentation. Although your interface should be intuitive, provide help and task-oriented instructions.
Usability testing. Perhaps the crown jewel of UX research is usability testing. This is where designers’ creations come to die. But also where valuable insights are generated and where the money gets made. Instead of me droning on about this, I’ll let a professional take over. Check out Nielsen’s five components of usability here. Click around to the different links. It takes about 10 minutes to get the gist of usability and usability testing, and plenty of experience to perfect the art. To give you a taste of what you’ll learn:
- Test early and often — all you need is a paper prototype.
- Test in context. Go to where your users are.
- Don’t interrupt the user or give clues.
- Ask what the user is thinking.
Goal: Optimize your site for meeting your business goals.
Once you release your design into the real, scary world, you’ll start to get visitors. That means data. You’ll want to check out the analytics of your site to see if people are truly clicking where they said they would, and more importantly, if those clicks are making your client money.
In order to optimize, you must have a definition of “more optimized.” For example, perhaps you want your conversion rate at or above 40%. It might not necessarily be the UX team coming up with these goals, but it is certainly your job to aid in the quest to reach the target.
Google analytics. Either you use Google analytics for analysis, or you’re a dweeb. Some things you want to look at:
- Total number of visits.
- Total number of page views.
- Average number of pages per visit.
- Bounce rate (percentage of people who leave after viewing one page).
- Average time on site
- Percentage of new visitors.
Again, the importance of these figures will depend on the goals you’ve set.
A/B Testing/Split Testing. If you’re not meeting your goals, try split testing. The idea here is to try different variations of your design to see which one works best. This is also known as A/B testing, where one variation is the “A” version and another is the “B” version (but there’s no reason you can’t try a C, D, E, etc. version as well). The variation that moves you closer to your predetermined goals wins.
Keep in mind that your sample size matters. Smaller changes require a larger amount of traffic to confirm efficacy. Brush up on your statistics or ask the smarty-pants on your team if you’re unsure of what I’m talking about. Further, be aware that you are testing variations live, meaning this could cause an inconsistent experience for actual users.
Doing your research
By now you’ve gotten a taste of what kinds of research are out there. It’s surely a lot to take in. But don’t worry. You’ll be great you little designer you.
In addition to checking back with those five aspects of good design research we talked about many moons ago, I want to give you a few other pointers to keep in mind while you’re actually doing your research. These will help you stay organized and make your research as conducive as possible to proper analysis.
- Show your work. Record everything. You need to be able to point to evidence that your insights and recommendations make sense. Or else you’re going to get the furry eyeball from the unfaithful. As in, those who don’t believe your research are valuable. Remember that the note taker is separate from the interviewer — you need to make sure you get everything down. Also, record research sessions either via audio (preferred) or video.
- Share your research artifacts. As I’ve mentioned, good design research is a collaborative process. Ensure that others can see what you’ve been up to and get good at showing your work. Get your own folder on a shared drive. Dropbox or Google Drive work well if your firm doesn’t already have the infrastructure.
- Audio is great, video not so much. This is a general statement. Of course, for usability tests, video might be the only appropriate medium. However, having said that, lean towards audio recordings instead of video recordings. Why? Two reasons: first, video takes up a ton of storage space; and second, video requires a hefty dose of editing to make it watchable.
- Invite stakeholders. Got an interview with a user? Invite a stakeholder! It will make your life so much easier if the stakeholder can hear a user state her problems first-hand. That way, there’s no room for suspicion when you speak on the user’s behalf.
Ah, yes. Ethics. That hollow buzzword you use to assure your boss you’re legit. (It’s cool — your boss does the same thing. Except his buzzword is “innovation” and his audience is investors.)
While conducting your research, you need to remember that you are prodding in the personal lives of others. Knowing how to conduct your work ethically is the difference between shaking your client’s hand for a job well done and shaking a lawyer’s hand to tell him you’d be happy if he could represent you in a lawsuit.
Calm down — you probably won’t be seeing your local attorney. Just make sure you consider the following before conducting your research.
- Gut-check. Take a look at the questions you want to ask. Would you be comfortable if you were being asked these questions? If so, then don’t ask. Take a look at the methods you want to use. Would you be comfortable in that situation? If so, then don’t do it. It’s really that simple.
- Consent and transparency. Don’t. Hit. Record. Until you get consent, that is. Make sure your participants know what your goals are for the study and how their information will be recorded and shared. If they say “no way,” then walk away.
- Safety and privacy. Make sure your participants know what they’re getting into when they sign up. You want to make them as prepared and comfortable as possible. And if you’re observing someone on his or her turf, make sure your presence doesn’t pose a risk. Like, if you’re observing a crane operator…you know, probably don’t distract him.
How to handle pushback
You will inevitably have haters that doubt the value of research. Instead of snapping back at them with distasteful glares and passive aggressive comments, fight back by dropping knowledge. As nearly all research projects require a certain level of organizational buy-in, get used to defending your work.
Things you might hear, and how to respond to them:
- We don’t have enough time/money. A classic retort of the research naysayers. You can’t really blame them — they’re probably under pressure from someone higher up than them to get results quickly and cheaply. They’re scared they won’t deliver. Don’t hesitate to play to that fear. Ask them if they have the time or money to be wrong about their assumptions. Tell them that even a little bit of user research and usability testing will save a complete debacle during the product launch.
- We don’t have the expertise. Horse hockey! This post alone provides a number of research methods so easy even a caveman can do it (sorry, cavemen. And sorry, reader, for the cheap Geico reference). Again, conducting research, no matter how elementary, will help you avoid disaster.
- We already know the problem/product/users. Does what you know answer your current questions specifically and shed light on how to accomplish your goals? And was this knowledge acquired recently? Erika Hall has a great quote in her book about this, “Familiarity breeds assumptions and blind spots.” Double-check your facts. Worst-case scenario you learn that you were wrong about what you knew and you avoided a complete disaster.
Analyzing your research
Phew, you’ve come a long way! You identified your research questions, planned the research, actually did the research…and now what?
The research you collect cannot help create good design unless you give it meaning. In other words, you must analyze your data.
Remember, I’ve said repeatedly that research is collaborative. The analysis phase is no different. What you want to do is gather your team (and stakeholders, if you can convince them) into a room with:
- Sticky notes
- Writing utensils
- Your research
- An open mind
Crack a smile for me — this is the fun part! We humans are great at this analysis stuff because we are hard-wired to look for patterns. We’re also social, so talk to your team mates as you go. Here’s a few activities to do:
Affinity maps. Read or listen to your interviews and jot down any sort of insight on a sticky note. It can be anything from “he is annoyed when he can’t put his man bun up perfectly” to “he doesn’t follow any man bun blogs.” As you go, patterns will start to emerge (that’s the power of your brain, baby!)
Once you have a bunch of sticky notes, place them on the whiteboard or wall. Rearrange them so that they’re in categories or groups. Now you’ve got some general insights.
Finally, write down some actionable things that your team can do to design for the insights you’ve found. For example, if you find that guys with man buns spend little time looking for man bun products, perhaps you’ll consider a very minimalist site.
User personas. Could you crush the role of one of your users in a movie? If so, you’ve gotten to know them well. Capture this knowledge in what is called a user persona. User personas are a descriptive profile of a specific type of user for your interface that captures the user’s goals, behaviors, pain points, demographic information, hobbies and technical literacy.
Once again: The biggest failure of the design process is designing for just one person: yourself. The greatest success is designing for those who you have sought to understand.
When you go off to design, check back in regularly with your personas. Let’s say you have a persona with the fictional name of Greg. Ask yourself when you create something: “Would Greg find this useful?”
The research don’t stop
So I’m running out of stuff to say about research. And perhaps you think that means that your research will at some point come to a definitive end.
Well, not really.
Teams that create great products don’t wait for a disaster to go back to the drawing board. Research is never finished. There will always be room for more usability tests, more interaction with your users and more analysis.
Enough of me talking, get out there and start researching!
- Just Enough Research — Erika Hall (or her shorter Medium post)
- User Testing’s Free UX Research Methodology eBook
- What and Why of User Research from Usability.gov
- Complete Beginner’s Guide to UX Research from UX Booth
- SWOT Analysis
- Competitive Audit
- How to do surveys
- Conducting Interviews
- Contextual Inquiry
- Heuristic Analysis
- Google Analytics
- A/B Testing
- User Personas
If you liked this article, please click the little green heart below — it don’t cost nothin’!
My name is Spencer Ivey. I’m pursuing a career as a UX designer and I just happen to love writing. Please check out my website! Peace and blessings.