Getting started with UX Research

This post first appeared in LA UX Meetup December 2016 - January 2017 newsletter.

As user experience professionals, we all realize the importance of getting real insights from real users and not just making decisions based on a hunch. So what can you do if you’re a designer who doesn’t have trained researchers on your team and you want to go beyond throwing your prototype in front of a few friends?

Well, here’s some help to get you started!

Step 1: Focus on usability testing

A well-designed, practical, usability study can tell you how users respond to your design and give you plenty of input on how to improve it, whether your design is a low fidelity prototype or a fully functioning product. Focus on usability testing and leave the interviews, surveys, and other techniques to the pro researchers for now.

Step 2: Create a test plan

I firmly believe anybody can learn to run a usability study, but unfortunately without preparation the study isn’t likely to provide the answers you need. Prep doesn’t have to be a formal process - you can keep this really simple and the whole study can be done in few days. First, think through three key questions:

  • What are you hoping to learn from the test (the objectives)? Your objective might be to see what issues participants encounter with your site when trying to find recipes they can cook for their family, or what keeps them from finding a car they may want to lease.

  • Who is your target audience and how will you find people like them? Are you more interested in professional chefs or stay-at-home moms? People who work full-time and need to make a quick weeknight dinner? And how will you find those people? Can you recruit them from your site with a tool like Ethnio, find them through a local cooking Meetup, or at a mommy and me yoga class? Recruiting the target audience may take some time and effort, so if finding those people is impossible then run the study with anyone who approximates them. It’s best to NOT use direct friends and family - people 1 or 2 levels removed from you is fine.

  • What will you ask the participants to do (what tasks will you give them) to address your objectives? It’s important that you GIVE PARTICIPANTS SOMETHING TO DO with your site and NOT ASK THEM HOW THEY FEEL or whether they like it. A great task would be “Find a recipe you can cook for dinner tonight” or “Find a price for a car that has the features you want.” People are lousy at predicting what they would do so don’t bother asking them that, just see whether or not they can complete the task you’re giving them.

Step 3: Don't go it alone

Before you go too far, it’s ESSENTIAL that you involve the rest of the team, which depending on your company might mean a developer, a product manager, someone from marketing, other designers, etc. Include these key stakeholders throughout the process as a way to get everyone to agree on how you’ll run the test so they’ll be more likely to accept the results.

Step 4: Confirm alignment with stakeholders

Write a short test plan to communicate the details of the test. This step is crucial to get all stakeholders aligned. Don’t worry about making it fancy - the plan could be literally a single page. Include in the plan all the things you’ve just worked out with the team: the objectives, the target audience and how will you find participants, and the tasks you give them to do.

divide.png

So now that you’ve identified the objectives of the study, your target participants, and what tasks you’ll give participants during the test, you’re ready to run the study!  

STEP 5: GREET PARTICIPANTS

You’ll need 5 or 6 participants for your study. Schedule each person individually (focus groups are the F-word of user research) and arrange a quiet space where you can meet. Once you’re with the participant, don’t forget to be personable and put your participant at ease. Greet the person, thank them for coming in, and chat for a moment to help them relax. Introduce the study by saying “I’m going to show you our site and get your input. There are no wrong answers, and please be honest - you won’t hurt my feelings. I’d like you to narrate your thoughts out loud as you work through the site so I know what’s on your mind.”

STEP 6: RUN THE STUDY

There are two mistakes I see people commonly make when they first start running their own research: they talk too much and they ask leading questions. The problem with both is that you bias your results by planting ideas in the participant's mind that they would not have come up with on their own. If you conquer these (bad) habits your results will be infinitely more useful.

Instead of asking leading questions (“Do you like the red button more than the green one?”) ask open-ended questions (“Tell me more about your response to the red button.”). Instead of talking too much, just give the participant the task (“Find a recipe you want to make for dinner tonight”) then stay silent, watch what the person does with your site, and wait for the person to talk. Asking “tell me more” is a great way to bring out more feedback in a non-leading, non-threatening way.

Get one of your key stakeholders to take notes while you run the study, just ask them to remain silent just like you will be. At the end you can invite the notetaker to ask any questions of the participant. Don’t forget to tell the stakeholders about the non-leading question thing.

Most importantly, don’t forget to relax! Running the study is the most fun part of the process so allow yourself to enjoy it.  

STEP 7: GOING FROM RAW DATA TO INSIGHTS

I think about data at many levels: there’s the raw data (e.g., “participant A could not find the Join button”), the trends that come out of that (e.g., “most people couldn’t find the Join button”), and the insights that come from those trends (e.g., “the Join button needs to be more prominent”).

First, get your stakeholders in a room and determine what you saw. The easiest ways to organize insights is to look at each task separately. What did you learn from all 5 participants about how they found recipes? Then go onto the next task.  

Once you’ve got the insights, I suggest you prioritize them into levels from severe (people could not complete the task, you really need to address this issue) to irritant (people were mildly annoyed), so you can easily identify the things that need to be fixed right now vs. later or not at all.

Then you’re ready to think about possible solutions to address the insights you identified. Caution: don’t confuse insights with solutions. An insight would be something like “The Join button needs to be more prominent” and a solution would be “Make the Join button red.”

And there you have it! You’ve just successfully run your first usability study. From here you can continue to deepen your usability testing skills.

STEP 8: HONE YOUR SKILLS

Getting a quick intro is great to start but ongoing mentoring and feedback is key to really deepening your skills. There is SO much info available, from online courses to books to in-person courses. Here are some to get you started.

  • Steve Krug’s books Don’t Make Me Think and Rocket Surgery Made Easy are classics for a reason. These are a great place to start.
  • If you search “User Experience Research” on Medium you’ll find posts by awesome UX leaders around the globe.
  • User Research for Everyone. Set of 8 talks by leaders in UX Research.
  • Observing the User Experience. The book of everything research, with detailed instructions on setting up and running usability studies and every other kind of study.
  • If you really want to get serious about adding research to your skillset, I suggest you take a longer term course so you really get to practice. For example in Los Angeles, Santa Monica CollegeUCLA Extension, and CSU Fullerton all have courses in UX research that will give you an opportunity to learn and practice techniques with a mentor. Look for a similar course in your city.