A Primer On User Research
By Jodie Moule
Good designers understand how to solve problems and create elegant solutions, but also know the value of considering other perspectives when doing so. Expertise within a given profession is often a combination of intuition, experience, some guesswork, and perhaps a touch of magic. Given the gulf between experience and more random factors, it’s necessary to move into your user’s world to gain a sense of how they live and work.
For the best outcome, you’ll need some level of user research in your design project.
Users Aren’t Designers
It’s unrealistic to think, however, that asking a bunch of people about your design problems will yield a complete solution.
If you’re a fan of The Simpsons, you may remember The Homer car. This is the episode where Homer’s half-brother Herb gives him a job at Powell Motors. Despite complaints from his staff, Herb encourages Homer to follow his instincts and create a car that average American consumers would want to buy. The outcome? Disastrous … and hysterical!
As well as being strange, Homer’s creation cost so much to develop and had such a high price tag that Herb’s company went out of business. In effect, Homer created a car that came out of his current life experience. This approach clearly failed to follow the rules of balance, whereby you should always account for desirability (human requirement) against viability (business requirement) and feasibility (technology requirement).
Never expect people to look beyond their current experience in order to come up with a great solution for you. That’s your job.
If we let our users make all the decisions (desirability without viability or feasibility), we’d end up with a horse designed by committee, as shown in Figure 1.
Figure 1. Never ask your users to design—you might not like the result.
Combining Methods for Best Effect
The most successful UX research projects involve participants who think carefully about the questions presented, enabling you to creatively plan a mix of methods that aid your design ideas. You should try to use at least two or three methods across a project life cycle, as illustrated in Figure 2. You can then review the results obtained across methods and observe themes, overlaps, or contradictions, as well as prevent misleading findings. It also allows you to determine the strongest insights from each method, resulting in a far richer understanding of your users.
Figure 2. Use multiple methods.
Choosing Your UX Method
It’s important to have a clear plan before you approach your users. Research is good for helping you with problem-solving and design work; however, research without a focus is merely guesswork.
UX methods are generally focused around observation as the data source, with conclusions being qualitative in nature. You can divide research into whether it requires direct or indirect user contact. The following table provides a summary of the different types of user research methods you might encounter, but it’s by no means exhaustive.
Different types of user research
Type of ApproachResearch MethodHelps to Answer …How many samples needed?Users’ preferences: their opinions, likes, and desires
- Questionnaire or survey (indirect)
- Focus group
- Customer feedback
What do users think about?Larger volumes: hundreds to thousands for a survey; four to 20 for focus groupsExploratory: investigating the context in which users complete tasks, helping to understand habits, motivators, drivers, and behaviours
- Contextual inquiry
- Behavioural interview
- Diary study (indirect)
- Mental models
What is an individual trying to accomplish?Smaller samples with this type of research: 12 inquiries or interviews; four to six diary studiesSummative: what is understood or accomplished with a product
- Usability tests
- Card sort (can be direct or indirect)
- Design workshop
Can users complete tasks?Smaller samples with this type of research: eight to 12 users for testing and card sorts; four to six for workshops
Different techniques will help at different stages of the UX process, so you should always choose the method that’s most appropriate for the types of questions you have to ask, as well as the needs of the project at that point. You may feel more comfortable implementing some methods over others due to their familiarity; for example, usability testing is probably better known to many of you than contextual inquiry.
Still, I encourage you to try each of them, as it will help you develop sound judgment. Over time, you’ll come to know which tool you should be pulling out of your research arsenal, and when best to use it.
A Festive Treat … Just for You!
As a special treat—and only for the month of December—the kind folks at SitePoint have offered UXmas readers the following discount:
- You can get the print/eBook bundle for $29 (that’s $39 off the usual price)
- Or, you can get the eBook for $13 (that’s $16 off the usual price)
Merry UXmas to you all—Ho, ho, ho!