What, So What, Now What?

The user research process is about going from observations to findings to actions. Step-by-step. In that order. This is obvious to long-time researchers. It’s surprisingly invisible to people new to user research.

For example, when we take notes in a research session we only note down observations. But if you ask someone new to research to take notes they’ll write a mixture of observations (participant struggled to enter a password), findings (hiding passwords stops users logging in) and actions (show the password!).

Mixing up these things causes us to jump to solutions too quickly. Or come up with general findings from too few observations. These things waste our time later.

I use What, So What, Now What to explain the user research process. It's about going through three stages, one at a time, pausing to reflect and prioritise after each stage:

• What?
  What happened? What did you notice? What facts or observations stood out?
• So What?
  Why is that important? What patterns, conclusions or hypotheses are emerging?
• Now What?
  What actions make sense?

What, So What, Now What is lifted directly from a facilitation technique used to debrief and wrap-up up workshops and meetings. It also happens to be an excellent way of thinking and talking about the user research process.

Cutting through our research jargon

One reason I love What, So What, Now What is that it helps people understand all the different, abstract words that we use in the user research process.

• What?
  Observations. Quotes. Videos. Photos. Facts. Notes. Clips. Transcripts.
• So What?
  Findings. Insights. Hypotheses. Theories. Reckons. Meanings. Explanations.
• Now What?
  Actions. Recommendations. Solutions. Next steps. User stories. Fixes. Designs.

Look at how many ways we have of talking about the same things! I'm sure you could add more. It's much clearer to talk about this stuff using What, So What, Now What.

Helping us prioritise effectively

Another reason I love What, So What, Now What is it shows that there are two different prioritisations that matter in the user research process.

The obvious one is prioritising the actions in the Now What stage. This prioritisation determines what work we do as a result of user research. It tends to be done by designers and product managers. We pay it lots of attention. It informs our backlogs.

Less obvious, but more important, is prioritising the findings in the So What stage. When teams jump ahead to solutions without doing this they come up with actions that fix unimportant or irrelevant problems.

People will always want to jump ahead to solutions. That's natural. It's our job to stop them doing this but, honestly, it's hard to explain why this is a bad idea. What, So What, Now What is a powerful explanation to arm yourself with for these discussions. 

Clarifying what researchers and designers do

The biggest reason I love What, So What, Now What is that it helps me explain the relationship of researchers and designers. When you split these roles it’s important to be clear about who does what in the user research process. How we work together. 

Here’s my take on how these relationships work in a perfect world. 

• What?
  Owned by researchers. We do research that generates good observations. 
• So What?
  Owned jointly by both of us. We create and prioritise findings together.
• Now What?
  Owned by designers. They decide on actions to address the research findings. 

Collaborating on the middle bit helps us get to actionable insights. The holy grail of user research. Researchers bring their understanding of human behaviour. Designers bring their knowledge of design problems. We do beautiful things together.

The big lesson for researchers new to product teams - especially those of us who used to work in agencies - is that we shouldn't usually be making recommendations. That's not our job, we're not best placed to do that, and it steps on others' toes.

Supporting less experienced researchers

In reality we don’t always work in a perfect world. There is often a mismatch in experience between researchers and designers on teams.

Let’s start with my tribe. Less experienced researchers struggle to create actionable insights. We report things that have little relevance or significance. Designers don’t know what actions to take as a result of the research findings.

In situations like these, I encourage experienced designers to take a leading role in shaping the findings during the So What stage. This can help researchers learn how to create actionable insights in future. It takes time to learn how to frame these things.

New researchers sometimes need help earlier too, in the What stage, because they’re not able to plan and do research that leads to good observations. Experienced designers often have good ideas about how to improve this research. I encourage them to share these ideas as suggestions.

Supporting less experienced designers

Less experienced designers struggle with actionable insights too. They often focus on findings they can imagine solving and deprioritise the rest. Big things can get brushed under the carpet because it’s not clear how to address them.

Here, I encourage experienced researchers to take a leading role in prioritising the findings in the So What stage. This can help designers learn to live with important findings that have no easy solution. This takes practice.

New designers sometimes need help later too, in the Now What stage, when the actions they take fail to address the research findings effectively. Many experienced researchers have good ideas about how to solve these problems. I encourage them to share these ideas as recommendations with the designer.

Adapting this to our everyday world

OK. These examples are a bit black and white. Let me cover some of the grey areas.

One is that it’s not just researchers and designers. Product managers, developers, and other team members are involved. What, So What, Now What flexes to include these people too because mostly (not always!) they act like designers here, in the sense that they're mostly interested in creating actions. Our teams love action.

Another is the idea of ownership. I said that researchers ’own’ one bit and designers ‘own’ another. Ownership is important for clarity of roles, yes, but user research is a team sport and so is design (at least in the sense of deciding which actions to take). I expect ‘owners’ to collaborate. In mature, experienced teams with high levels of trust I'd expect to see researchers and designers collaborating at every stage. 

The final is that these things are fluid, contextual and situational. No one way of working will work at all times. Even on the same team with the same people. It is up to researchers, designers and teams to make What, So What, Now What work for them. Let me end with an example that illustrates this from my own career...

Ending with an example from GOV.UK Notify

I worked as a user researcher on GOV.UK Notify with exceptional designers and product managers who made great decisions based on our research. 

There was one time when their actions were not addressing the research findings. We saw the same problems again and again. I agonised about intervening because I’d been clear that my role ended with creating and prioritising the findings together.

Eventually I chose to act. I explained that I was going to step, briefly and exceptionally, into Now What territory and suggest actions to solve the problems. I told them they could ignore these. But they listened and our team moved forward.

This is why What, So What, Now What is so useful. It not only explains the user research process and clarifies our roles, it also gives us a framework to talk about the inevitable exceptions. It’s OK to break the rules if you know the rules in the first place. 

Try using What, So What, Now What the next time you explain the user research process to someone. I think it'll help you have clearer conversations.  

Let me know what you think on @myddelton. If you ever run workshops you should try the What, So What, Now What exercise from Liberating Structures. I have a suspicion this pattern works in more places than user research and workshops too…