How to Do UX Research

Most UX teams live every day with the anxiety that they don’t truly know what their users want and need. They think they know. They build features and experiences that they hope will satisfy user needs. They may even do some usability or user acceptance testing to work out the bugs and make sure what they built works as intended.

In our experience, though, most organizations don’t have the time and resources to do the type of in-depth research that ensures you’re solving the core problems of your users.

After all, what’s a product worth if it doesn’t help a real person solve a real need in the real world?

Over two blog posts, we’re answering two fundamental questions about UX research - why to do it and how to do it.

Part 2: How to do UX research the right way.

The only thing worse than not doing any UX research is doing the wrong UX research.

It’s hard enough to build a successful product for users when you don’t understand the problems you’re trying to solve for people. It’s a whole other challenge when you find you’ve been solving the wrong problems because your research led you down the wrong path.

Here are five steps you can take to make sure you create and execute a research plan that will deliver real results for your users, your customers, and your business.

Step 1: Clearly Define Your Objectives

The first step in putting together a research plan is to answer the most important question: Why are you doing this research?

When not thought out or planned well, user research can be expensive, frustrating, and time-consuming and it won’t end up delivering much for the effort.

You need to be able to clearly articulate what you’re going to gain by doing this research and what success looks like for you and your team.

Here are two ways to clearly define your goals and criteria for success:

1. Talk to the people who have a stake in the success of the product. We throw the term “stakeholders” around a little too loosely sometimes. Think of it literally and put sincere effort into clearly understanding the stakes. How can research help those stakeholders answer their most important questions? What would success look like for them and for their business?

2. Review before you start. Too many research projects rush past this initial research step in anticipation of generating new insights. Previous research studies, existing analytics data, customer support tickets, and competitors’ products are all great sources of information that can help you more clearly see what you already know and what you need research to help you uncover.

Step 2: Choose your Methods

Based on the information you gathered when defining your objectives, your next step is to determine the best research methods to uncover the answers you need.

There is one key question to ask when choosing your research methods.

Do you want to hear what people say about something or do you want to see what they actually do?

If you’re interested in what people say or think about a problem or potential solution then you may want to start with a quantitative survey followed by user interviews. A quantitative survey will measure and categorize a user’s stated opinions toward your product or service and identify areas where you can dive deeper in interviews. But they’re only as reliable as what people tell you they might do in a hypothetical future situation or what they remember doing in the past.

If your research questions are more focused on actual user behavior then you have a few options. You could choose to run a usability study that’s moderated by a researcher and recorded to track how users complete specific tasks. You can also try an ethnographic (or field) study that watches people “in the wild” as they go through the steps to accomplish a task or solve a problem your product is designed to fix.

Either way, it’s a good idea to mix quantitative and qualitative methods so you get a good picture of what people are saying side-by-side with what they actually do. Turns out, we’re all human and we often say one thing while doing the exact opposite.

Pro Tip: Video clips are often the best way to get stakeholders to see the issues and feel the users struggle in specific contexts.

Step 3: Recruit the Right Users

No matter what research method you choose, you’ll need people to participate in the research. But it can’t be just anybody you know who “kind of” fits the profile. You need to source real users who really do (or would) use your product or service.

This is often the most time-consuming part of any research project because it’s a little like herding cats. People don’t show up for scheduled sessions, change their minds about participating or just don’t feel interested in giving you feedback.

To make recruitment as effective as possible look at your current user base or analytics data to choose people who fit the right profile or behavior pattern for the questions you want to answer. If you haven’t launched yet, do some market research and look at who uses similar products in your space. Recruiting the right people means you can talk to less of them because you’ll quickly start to see the biggest trends that are most common to your core user and you can more thoroughly understand and empathize with their pains.

Recruitment agencies or online tools like UserZoom, User Interviews, or Respondent.io are good sources for recruiting users quickly. However, they can get expensive, don’t always have the user you need, and can deliver folks who are a bit more experienced at testing products than an average person (the dreaded professional testers). You may need to do some guerilla-style recruitment through Facebook groups or personal and professional networks.

One of the best options if you have a large existing user base and are testing things like usability of new features is to cultivate a user feedback program or panel. Your existing user base can sign up to be on-call research testers that you can tap into from time to time for feedback on new features or ideas.

However you source your participants, you’ll need to think about compensation. It’s extremely rare to find a lot of people who will give you their time and feedback for free. Compensation needs to be based on the type of participant and their role. Professional users responding to a research request for a tool or process related to their work will expect something close to what they make in their salary or hourly wage. Users opening their personal lives to research may respond to slightly lower compensation but it still must match their general expectations for what they would want to receive in exchange for the amount of time and effort you’re requesting.

Step 4: Create a Research Plan and Scripts

This is another step that’s often overlooked in the rush to get in front of users and gather quick feedback and insights. But a good research plan will tie together your research questions, the methods you’ve chosen to answer those questions and your recruitment plan into a cohesive outline that you can follow as you conduct the research and gather the data.

Scripts help researchers stay focused and on track and remember to cover all the research questions and topics. They also allow multiple researchers to moderate different sessions with consistency. Qualitative, open-ended questions help get the user talking about how they currently accomplish a task or solve certain problems related to the product in their day-to-day lives.

The key to conducting great research is having a great plan, putting the plan into action, and then shutting up and listening. It’s easy to overreach and ask too many questions or direct the user toward your own assumptions. The most powerful tool in user research is silence. Users will often fill the silence with an insight or observation you never would have thought to ask about.

Step 5: Launch Your Research

With a plan in place, it’s time to schedule your research sessions and execute the research plan. The key to this step is not being afraid to make mistakes and adjust on the fly. No two user interviews, usability sessions, or ethnographic studies are the same. Just like your product, your research plan needs the flexibility to adapt to real users. You’ll learn a lot from your first few sessions about what works and what doesn’t. Adjust your plan as needed but keep your focus on the main questions you’re trying to answer.

A few additional tips:

• Try to remain as neutral as possible to prevent influencing users during your research.
• Don’t ask leading questions (this can lead to confirmation bias of your own assumptions).
• Do ask open-ended questions (this can lead to new and unexpected insights).
• Try to stick to the script and avoid answering any questions from users that may prevent you from witnessing a valid and helpful pain point in the process.
• Try to gain as much context into a user’s actual behaviors as opposed to their intended behaviors or what they say they would do in an ideal situation.
• Invite other team members and stakeholders to observe research sessions for themselves. Nothing brings the point home better than seeing a problem in real life.

Looking for a Research Partner?

We’re experienced in research methods of all shapes and sizes, from ethnographic studies to quick and agile usability testing. We pride ourselves on choosing methods and building research plans that drive real results.

Start a conversation and see how we can help take your user research to the next level.