Good UX design is built on good user research. User research is the best way, indeed in my opinion the only way to really understand your users and to build the foundations for great UX design. Without user research you’re guessing, it’s as simple as that, and as Mark Twain reminded us long before UX exploded into the Universe, finding out beats guessing every time.
You know you definitely need to carry out some user research but what? How? Where? With whom? As with so many things in life good user research requires careful forethought and planning. Planning always makes my head hurt so to make the process as quick and as painless as possible here are are 10 questions that I always like to ask before undertaking any user research.
1. Why is the research required?
The most important question of them all – Why is the research required in the first place? Looking at the background first can help to provide the context for the user research and reveal how it fits into the bigger picture. It might even indicate that the sort of user research you had in mind isn’t even necessary in the first place.
2. Who has a stake in the research?
It’s important to think about who the key stakeholders are for the user research and who the findings will need to be communicated to. Stakeholders might include business stakeholders, developers, designers, project managers and so on. These are the people that will need to be directly or indirectly involved in the research, and who will more than likely need to be won round to ensure that user insights are actually used, rather than simply gathering dust in some desk drawer somewhere.
A good way to determine stakeholders is to use a quick responsibility assignment matrix (RACI). Simply list:
- Who is responsible for the research
- Who is accountable
- Who should be consulted
- Who should be informed
3. What are the goals of the research?
User research should have clear goals, otherwise it’s really just research for research’s sake. Think about what the key goals of the research are, and how you will know if the research has been a success or not. Sometimes it’s a good idea to work backwards from a goal. Take a goal and think about what you will need to discover to realise that goal. You can then think about the research methods you can use to make these discoveries, and the type of users you will need to involve.
4. What questions are you trying to answer?
User research is really just an exercise in answering questions. Who are our users? What do they do? Why do they do what they do? What do they need? Planning user research can sometimes feel like having a conservation with a 3 year old, like a never-ending tirade of question, followed by question, followed by question. There are an infinite number of possible questions out there so stick to key questions and think about what you really need to find out to be able to move forwards.
5. What assumptions and hypotheses need to be checked?
Assumptions are not always a bad thing. Assumptions let you move forward, but like a badly worn tire, left unchecked they can all too easily blow up in your face. Think about what assumptions you’ve made that you will need to check. Also think about any hypotheses you’ve made that will need to be verified. For example, that a particular type of user is most important, or that a product or service is used in a particular way.
6. Which research methods and tools will be most effective?
A key question for any piece of user research is to determine the research methods and tools that will be most effective. Your choice should be influenced by the level of access you have to users, the type of data you want to capture and even your own experience and comfort with a method or tool. As a general rule qualitative methods will tell you more than quantitative, and face-to-face will tell you more than remote. However, this is one instance where ‘it depends’ really does hold true. The most effective research methods and tools will undoubtedly change from project to project.
7. Who should take part in the research?
User research is only as good as the users you conduct the research with. Get the wrong users, and you’ll be collating the wrong user insights: Garbage in = garbage out. You therefore need to carefully consider who should take part in the research. Who are your users and which users groups will you need to include? Qualitative user research can often be carried out with as few as 5-8 users, but you’ll need a lot more than that if you have a very diverse user base, or of course if you’re collecting quantitative data.
8. Who should carry out the research?
User research is a bit like cooking. With a little bit of training and practice everyone can do it, but it takes a lot of skill and experience to do it really well. It’s important to think about who will be carrying out the research. Is there expertise available in-house or is it best carried out by a 3rd party? I would always aim for in-house researchers, rather than 3rd party because this helps to keep the knowledge and insights within the team, and makes it more likely that the research findings will be actioned. However, if expertise doesn’t exist in-house, if time and resources are not available, or perhaps if the research has all the hallmarks of a political hot-potato, then using a 3rd party can often be the best way to go.
9. What should the research outputs be?
Doing the user research is very often only half the job. Actually doing something with the insights is the other, much more important half of the job. It’s important to think about what the outputs of the research should be, and how outputs will best be communicated to stakeholders. Rather than spending ages crafting the mother of all research reports I always try to keep outputs as lean as possible. Very often a quick talk-through and if necessary presentation of the key findings is all that is required. Videos and visual presentations are especially good – I particularly like Nick Bowmast’s advice for visualising design research.
10. When does the research need to be carried out?
It’s important to think about when the research will need to be carried out by, and to look at how much time there is available for the user research. This will very often dictate what is and isn’t feasible. More often than not there will be less time available than you’d like, and with more and more projects being carried out in an agile fashion, research time is being increasingly squeezed. Even if there is very little time available, it’s always worth carrying out some quick user research. After all, a little quick and dirty user research is better than none at all.
A user research canvas
I’m a big fan of using a canvas, like the Business Model Canvas, or indeed my own Service Model Canvas to help ensure that the information required has been discussed and captured. As you can never have too much of a good thing (at least not until the next morning when the headache kicks in) I’ve created a user research canvas to help discuss and collate the information covered by these key questions. You can download the canvas using the links below and either fill it out yourself or with others as part of a rapid fire user research planning workshop (RFURPW for short). Good luck with the planning!
View full size user research canvas
Source: UX for the Masses
