There are lots of different situations that can provide the impetus to begin UX research. Companies who are moving into a new market, launching a new product, extending a product line, or who have watched a project tank for unexpected reasons (not to mention companies who are in the founding stages) - all have something important in common: it’s abundantly clear to them that at these moments in time, critical pieces of knowledge about an audience are missing. And, more often than not, the answers seem truly elusive - none of the current data-gathering methods seem to provide the needed information.
More and more teams now turn to qualitative user research to solve this problem, either hiring a specialist, an agency, or in some cases, equipping their in-house product design team to lead the effort. UX research, an important subset of user research, will make up a large part of your user research efforts if your product relies on a digital interface, but the basic work processes are the same. If you are considering making use of UX or user research, here’s what you need to know in order to make the most of it:
The question is king
We spend most of our workday trying to convince other people that we know what we are doing, that we have the right answers, that our point of view is the correct one, and that we are likely to succeed. But this is shortsighted, because it blinds you to the insights which lie outside your field of view, and are waiting to be discovered. It may feel strange at first, but there’s no real risk in admitting what you don’t know. When beginning research with clients, I let them know that this is the time to give me the opposite of their pitch. A conversation about research is the perfect time and place to admit to yourself what you really don’t know about your product, business model, or audience. Those kinds of questions that you’re afraid to admit to your investors, your boss, even to yourself - the questions that, when you ask them out loud, you’d fear make it seem that you might be incomptetent at your job. Write them down, and you can begin to get some answers.
Once you’re able to articulate your most important questions, a good researcher will then hone and focus them even further. This process does two things. Firstly, it boils down the questions into a focused hypothesis that can then be tested out quickly using methods that the researcher will recommend to you. And secondly, the researcher will help ensure that the question being asked does indeed reach the heart of the matter (or the heart of consumers) - ensuring that the insights you gain will be the most profitable ones you can get.
Mostly you’ll need to shut up and listen
The best researchers will make themselves almost invisible, setting things up (or in some cases, putting people at ease) so that research subjects will provide you with boatloads of unbiased data about your product. If interviews or focus groups are part of the process, you will often have the chance to observe the proceedings. You will be briefed beforehand to stay as quiet as possible and to jump in only at certain cues. It can be difficult for some people to do this, but you will be richly rewarded with more genuine perspectives from users. Most qualitative research methods are crafted around the idea people are testing out your product all the time - you just don’t get to see it. A carefully crafted interview or other research setup is there to help you cast a net and learn from what is happening out there in the wild. For newcomers to the process, it’s hard to underestimate how much of an impact it has on team members, or c-suite executives, to simply be present at a successful interview. I’ve heard words like “mind-blowing”, “fantastic” and “eye-opening” from clients. For many people, it opens a door to powers of observation that they didn’t even know they had.
Find evidence, then validate
A common concern about beginning the research process is the idea that a small number of people aren’t able to represent what the masses may be experiencing or doing. It’s easy to understand this concern. Depending on the research method chosen, your number of subjects may range from 5 to a few hundred - a very small subset of what may be tens or hundreds of millions of people in the target market. How will you know if these few subjects are a good stand-in? The answer to this question is twofold. In a world where quantitative data has become cheaper and more easily accessible than ever, it can be tempting to look at whatever relevant numbers are available to you, and assume that they contain the whole story. It is easy to forget that most of the time, those numbers are simply an answer to a question posed by someone with an existing bias. In the same way as described earlier, if the question wasn’t composed properly, then it’s likely that the numbers you’re getting are not giving you insight into the most important details of customer action, instead focusing on peripheral stand-ins for success. In these cases, the data is gleaned from more people, but it is shallow enough to be missing many important facets, that will emerge from qualitative research - such as what were people’s motivations to buy, install, or take other actions that are recorded in the data. This is the first half of the answer to the question of small numbers in research studies.
The second half of the answer is that qualitative research should never happen in a vacuum. It is often bookended with pretty straightforward quantitative analysis. Maybe you approached qualitative research because of discrepancies you observed in data available to you, and you want an explanation for that phenomenon. Maybe you noticed that your key metrics lag in one area and want to know how to improve. These are cases where qualitative research complements your numbers. Additionally, a common method of answering research questions conclusively is to use a two-step or three-step process of validation. In this process, you begin making informed guesses about a larger audience via insights you uncover with a very small number of interviewees or research participants (usually a minimum of 6). Then, studies are conducted with progressively larger groups to validate whether the experience of the few is indicative of the wider group. By the time you’ve reached a third round of research participants (say, via a survey to thousands of people) you’ve begun to reach significantly sized sample groups. Some might be tempted to skip straight to this larger survey, but it isn’t usually advisable: surveys rarely provide enough depth on their own. Whichever the method chosen, a good researcher will help make sure that you aren’t blindly pinning conclusions on too small of a group.
Success leads to more questions, which leads to more success, which leads to…
Finding insights in this way soon becomes addictive to teams, who use it to supercharge decision making, and often end up saving money as well as time. I recommend keeping a wiki or a log of new questions as they come up. You will find that as you or your team get the hang of things, everyone becomes better at asking the right questions and research cycles get shorter, with questions being answered in less time. Sometimes, incorporating a research process even helps with team dynamics: people begin to realize that they don’t need to argue their points - instead they can put them to the test.
Still have questions about how things work? Drop me a line here.