When I'm faced with a design problem that requires some research, I often look for the most efficient way to find answers.
Over the years, I've assembled a list of UX research methods with a high return on investment. I've found again and again that they give me a lot of bang for my buck: they don't take much time or money but are still high impact.
These are the go-to methods I can fall back on when I need to learn quickly and effectively.
But before we jump into the list, it's important to understand two things.
What exactly are we looking for?
When designers do research, we're looking for one main thing: insights.
Insights are the takeaways—the valuable nuggets, lessons learned, epiphanies—that help us discover new ideas that we can then go apply to our designs. Insights are the fruit of the research and the whole reason we do it.
So what we're looking for is the most insights per time and money spent. The better that ratio is, the more efficient the method is.
Can I use the same method every time?
Here's another important thing to keep in mind: There isn't one blanket research method that works for every single situation. Research and UX design in general is too complex for that. There are too many variables at play.
To figure out what method is most efficient, we need to understand what question we're asking. What do we need to learn? What's the goal? What problem are we trying to fix?
For example, are we looking for numbers (metrics) or stories (events)? Are we analyzing behavior or attitudes? Are we watching or asking? Are we testing something tangible or trying to get into the head of our user?
I’ve listed several methods below and each one is tied to a specific type of design challenge or goal. I've found that the associated method is almost always the most efficient way to get to the answers.
My Favorite Efficient Research Methods
1. User Interview
Goal: Discover mental models, attitudes, and opinions of your user on a specific topic.
In my mind, there's no replacement for sitting down with a user and digging into their mental models via well-crafted questions. You can gather many insights from this experience and it's very adaptable. There isn't another method that works as well as this one for tackling this specific type of challenge.
Interview Length
Interviews don't have to be hour-long calls—they can be a quick 15-minute chat. I find that designers sometimes make interviews too long just because they feel like they need to be long. Don't get distracted. Know beforehand what exact question you're trying to find the answer for, and write questions that get you to the answer as quickly as possible. Once you’ve found the answers you’re looking for, get off the call. There’s no reason to just hang out.
User interviews can easily be completely free or at least very inexpensive. Record it for free with Zoom or your device's default audio recorder. I've found that participants are usually willing to do an interview for free if it's quick. You can also offer them something that doesn't cost you anything, like a free month of your product.
2. Usability test
Goal: Evaluate a specific idea you already have.
Usability tests are incredibly effective: they give you tons of learning opportunities and insights. This is largely because you're testing tangible work rather than gathering general concepts to be used later. You can even fold in interview-like questions that help you get some of the benefits of user interviews. Tests give you concrete, actionable takeaways that you can immediately apply to real designs. They steer your work in the right direction and get you closer to shipping.
If I had to pick one method to use for the rest of my life, it would be a usability test. It has several of the same benefits as other methods on this list and is always very practical and actionable. You're not talking about abstract concepts, you're talking in terms of the actual things you're building. If you only have the budget (time and money) for one research method, you can't go wrong with a good usability test.
It's getting easier to do remote, unmoderated testing with tools like Maze and UserTesting. The main benefit of these tools is that you can set the test up, find participants and then let it run in the background. These can sometimes come with certain disadvantages like not being able to adapt the direction of the test based on the participant's responses. But I am growing to like these methods more and more lately and find that I can get quite a lot of insights out of them if I build them correctly.
3. Diary Study
Goal: Analyze a user's current behavior and relationship with the product.
Diary studies give you a great way to see the details of how a user uses a product on a day to day basis. They help you get a feel for the user's overall relationship with the product and how it fits into their life.
Note: Some diary studies can last for a while. It's in this list because you can set it up pretty easily and let it run in the background while you go work on other things. But usually, you want to let the test run for a couple of weeks; so be aware that you should only use this when you can set it up that far in advance.
Since diary studies don't happen within a small, controlled window, the user's real, natural behavior emerges over time. A longer duration also means you have a broader time period and range of situations to make sure you have a higher chance of getting all the problems and not missing the small gripes that pop up here and there when you're using a digital product. These types of things often don't emerge in a small-scope, quick-time-boxed test.
You don't need to go crazy with the tools. Use things like Google Forms or Notion.
4. Card sort
Goal: Build an information architecture.
Quick, simple, and often free to set up, card sorting is invaluable for figuring out how to structure something.
It's similar to usability tests in that it provides lots of opportunities to learn from the user and gives you concrete, actionable takeaways that you can immediately apply to real designs.
There are lots of tools out there (like Trello + Zoom) that will make this free and easy.
5. A/B (or multivariate) Test
Goal: Figure out which of your ideas will best meet your objective.
If you have a specific goal in mind (like you want the user to click on this button, or utilize this feature, etc), and you have a couple of options for how to go about it with no clear indication of which would work better, A/B or multivariate tests are invaluable. They'll give you direct, quantitative feedback that's pretty hard to refute. And it's nice to set it up and let it run in the background, gathering data.
The issue with A/B tests is that if you're going to do this in production, you need to have (a) the technical capability and (b) at least two things ready to ship. You can make this easier on yourself by only testing one small piece of UI (the scope of an A/B test should be as small as possible, anyway). And there are lots of tools these days to help you run one of these tests, like Google Optimize or VWO. Or you could replicate a production A/B test with an unmoderated usability test.
If you have the particular goal outlined above, this is the often quickest, best way to get to a clear, evidence-backed decision. Companies like Facebook and Google use this all the time for that very reason.
6. Survey
Goal: Determine the priority or scale of a problem among your userbase.
Need to figure out what proportion of your users share a common opinion or are impacted by a specific problem? Surveys are the best way to figure that out because they give you a wide scope. And the answers will be incredibly valuable for deciding what to build and how to build it.
You could also reference analytics to try to address this goal. Although, you'll likely have a harder time pulling out real insights because it's retroactive and doesn't directly address the question.
Writing surveys should be done thoughtfully, but after you've written them (and gathered the participants) you can send it out and let the responses come in. The other major benefit of surveys is that you can easily run them for $0. You can use a tool like Google Forms for free and then export the results to something like Airtable (also free).
Wrapping Up
On the UX Chats podcast, we talked with Trae Winterton about this very subject. Go check it out!
What methods do you default to when you need to find answers efficiently? Let us know on Twitter.
This article was first seen at: https://uxtools.co/blog/best-ux-research-methods-in-a-pinch/