Cookies on Knowhow

We use cookies in order for parts of NCVO Knowhow to work properly, and also to collect information about how you use the site. We use this information to improve the site and tailor our services to you. For more, see our page on privacy and data protection.


Skip to content. | Skip to navigation

Community-made content which you can improve Case study from our community

How to carry out effective user research

Building a new digital service can be hard. How do you know what features your users will need the most? Or what platform they’ll be most likely to use? Fortunately you don’t need to have all the answers yourself. In fact, you shouldn’t. Because that’s what user research is for. 

User research is not the same as market research, and 1:1 interviews will give you more valuable, in-depth insights than surveys and focus groups. The point of user research is to find out about your users’ attitudes, behaviour, goals and challenges. It’s about uncovering how they see the world, and how they behave in it. If you understand this, you can design and build a service that users’ need, not what you think they need.

Before starting any user research, think about the people you’re engaging with. Put yourself in their shoes. This will help you decide how best to interview your users and how to make them feel at ease in the session. Anything from cups of tea to ice-breakers can help build rapport and trust.

Tips for effective user research


Don’t just ask users what they want

It’s often unrealistic to ask your users to guess what product or feature is likely to best solve their problems because they aren’t technology designers. It’s helpful to understand people’s preferences, but user research should go beyond that. For example, if a user says they want a certain feature you can ask them, ‘what would that help you to do?’ Asking this means you can understand what problem the feature solves, and when it’s needed.

A second reason you don’t just ask what someone wants is because people aren’t able to articulate solutions they don’t know about yet. There’s a famous saying in user research circles called ‘faster horses’. It supposedly comes from Henry Ford who said that if he’d only listened to what people asked for, he would never have built a car. Instead he’d have created faster horses, because that’s the mode of transport people were used to, and that’s what they were limited to thinking about.


Don’t ask what someone will do

Similarly, asking someone if they will do something, or are likely to use a product or service, is unhelpful. People say they will do all sorts of things and may mean it at the time. But it’s not good evidence to build a product or service on. If you have a question in an interview that says, ‘would you be likely to do x in the future?’, scrap it. It’s more valuable to ask about their real past behaviours than their hypothetical future behaviours. But if you really want to know if people will do something, build a prototype and test it.


Observe and document users’ behaviours

Use different methods to uncover user’s problems, attitudes, goals and habits. For example, you might spend time observing how users work (called contextual inquiry). You could ask them to walk you through a task they do step-by-step or to complete a diary study, noting down things they do during certain times in the week. All of these aim to get as close as possible to documenting the activity as it happens.

If you only have users’ reported behaviour to go on, focus on specifics. For example, if you’re exploring people’s digital habits and can’t observe them directly, ask them to describe a particular time they used a specific device recently (eg in the last month). What did they use it for? Why did they choose that device?


Remember that recruiting users for interviews is always the biggest problem

Without fail, the biggest challenge is getting users to interview or test with. There are lots of ways you can do this, from putting out a call to service users in your network (or the gatekeepers to those service users) through to using a professional user-recruitment agency.

There are pros and cons with both approaches. Whatever you do, make sure you plan it plenty of time ahead. With charities with vulnerable or hard-to-reach users, it’s even more important to be patient and flexible.


Don’t ask leading questions

Leading questions massively bias your findings and should be avoided. Shouldn’t they? That said, it’s easy to catch yourself asking one when you’re in mid-flow during an interview. To avoid this, write out your questions in advance, including potential follow-on questions. Anything that starts ‘Do you think…’ or something similar, is almost certainly a leading question. Non-verbal cues can also be leading, so be conscious of your body language when interviewing people face to face.


Remain neutral

People are generally nice, so they often say things that they think you want to hear. Try to be neutral with your verbal and non-verbal responses. It doesn’t mean you have to be a robot, and it’s always important to make participants feel comfortable, but try not to steer them. This can be hard as normal conversation is often peppered with supportive agreement (or disagreement!). Being neutral starts from the moment you contact someone. Telling them too much about what you’re trying to do, or your thoughts on a topic can steer their responses before you even start an interview. 


Gather appropriate consent

Users have a right to know how you’ll use their data. And you have a legal requirement to manage it responsibly. It’s best to create a template form that’s relevant to your organisation, context and users.


Leave enough time for analysis

When you book out time for interviews or usability testing, book out time for the analysis as well. A good rule of thumb is to have at least the amount of time it took do the interview or testing session. Where possible, make analysis collaborative by involving the wider team, including in-house developers or the tech partner you’re working with. The findings will mean much more to the team if they understand where they came from.


Plan and track your research

Each interview or user testing session you run takes a lot of time and effort to prepare and analyse. So make sure you are planning your research and being clear about what questions you’re trying to answer in each round of interviews. This means you’re maximising value from your research activities. At CAST we track our findings and research questions using things like the knowledge kanban.


Communicate your research

One of the many things I learnt at the Government Digital Service (GDS) is that user researchers spend more time communicating research than doing it. It’s not much good doing lots of useful research if it doesn’t shape the direction of the product. So plan ahead about how to communicate your research to the team.

To get started, try thinking about:

Even better, get the whole team involved in research. Afterall, user research is a team sport!

Further information

This blog was written by CAST - Centre for Acceleration of Social Technology

These are of course just 10 tips - we’d welcome your additions to this, so that it becomes useful for the sector.



Page last edited Apr 23, 2019 History

Help us to improve this page – give us feedback.

1 star 2 stars 3 stars 4 stars 5 stars 2.9/5 from 313 ratings