Analysing insights

User research activities produce a lot of raw data, for example:

• written and digital notes
• sketches and photos
• audio and video recordings.

You need to filter, organise and interpret this data so you can produce useful insights that will help you design and deliver your service.

Plan analysis sessions

Invite anyone who observed the user research to take part in the analysis session. You should do analysis as soon as you can after each round of research, while it is still fresh in people's minds.

Involving lots of people helps your team make better decision. It reduces the risk of researcher bias and limits the influence of individual team members or stakeholders. It also means you can decide as a group what the next most important thing to work on should be.

Aim to spend one hour analysing per two hours of research.

Extract observations

Follow these steps to gather observations:

1. Hand out sticky notes (use just one colour).

2. Ask people to review their notes and write down anything interesting or relevant that they saw or heard during the session.

3. If you have recordings, make them available so people can confirm their observations and get verbatim quotes.

4. Tell the group to use a single note for each observation and write exactly what they saw or heard (eg verbatim quotes or observed behaviour), not what they think it means. This way, the notes will be unbiased and can represent the voice of the user.

Sort observations

Once they've written down their observations, ask your team members to place their sticky notes on a wall and start sorting them into similar themes, for example:

• common topics (for example, identity, delivery, payment)
• stages in a user journey (for example, 'provide photo', attend interview', 'pay')
• individual pages or steps in a transaction
• types of user (for example, personas)

Allow people to move notes placed by other people. The idea is to look for patterns or departments, and to keep moving the notes until clear groups emerge. This is often called 'affinity diagramming' or 'affinity sorting'.

Name your groups

Once you have your groups, agree a title for each. At this stage, you can discard irrelevant or isolated notes. You should also check if you can break any large groups into smaller themes based on matching observations.

For example, if users have to provide a photo to use your service, you might have a 'photos' group that could broken down into:

• photo rules and requirements
• using a photo booth or high street photographer
• taking a photo at home
• reasons a photo might be rejected

Determine findings

The final part of analysis involves reviewing the notes in each group to determine what the observations are telling you.

When you agree on what you've learned, write it as a finding or 'insight' on a different coloured sticky note and add it to the relevant group.

You should write findings as short statements that summarise what you've learned, for example:

• 'the legal declaration is threatening and difficult to understand'
• 'people think they can click the progress bar to navigate'
• 'users are confused about what's required because so many questions are optional'.

Decide actions

You can use your findings to make decisions about what to work on, change or research next. This supports the agile method of planning continuously based on new facts or requirements.

As a group, discuss if there are any obvious actions you want to take. Write these on a third colour sticky note and add it to the relevant group.

These might include:

• new design ideas to work on
• new questions to include in user research
• things you want to change in a prototype and test in another research session
• new user stories to add to the product backlog
• new details you need to add to an existing story
• strategic insights you can use to develop your user needs, proposition or product roadmap.

Case studies and examples

To learn more about analysing user research sessions:

Find out how GDS does research analysis in agile

See how to use sticky notes effectively when analysing research

Find out why user researchers at the Department for Work and Pensions rely on insights

Find out how to do large-scale group analysis of user research data