https://dwpdigital.blog.gov.uk/2015/03/26/digital-academy-the-importance-of-insights/

Digital Academy: The importance of ‘insights’

Ben Holliday - Head of User Experience Design
Ben Holliday - Head of User Experience Design

In my previous Digital Academy blog post, I wrote about how to get started with discovery research.

Ideally your team will have been part of this process, getting involved with research and taking part in analysis sessions.

For people like product managers or key stakeholders this won’t always happen. In agile teams time is limited. The reality is people won’t have the time to read research reports and they won’t always have the same shared understanding of problems when it comes to planning and prioritisation.

Teams need ways of getting hold of the actionable learnings from research. The key to this is ‘insights’.

Introducing insights

Insights are short statements that give us enough shared understanding to take an action. For example, this is an insight from our State Pension research:

“People confuse their current State Pension with the amount they’ll get when they retire”

The best ideas, innovations, or product iterations can come from a single insight like this. It’s worth considering that insights will often be the result of frustrations­ seeing people struggle with a product.

What makes a good insight

Insights should feel simple because they are simple. They should also be provocative – this is what makes them actionable. They should challenge how we think about problems we’re trying to solve – helping us to find new and improved ways of meeting user needs.

The best insights are like problem statements. We want our teams to use their experience and skills to solve the design challenges they create. Insights should make space for this to happen.

This is where the big and interesting ideas are. Designers will naturally make intuitive leaps. They’ll be able to develop ideas that move a team towards building something that works better for users – solutions that respond to individual insights.

User research won’t tell you what to do. You need to work towards solutions using what you’re learning as a guide.

Insights arent an exact science

It’s important to remember that we’re looking for insights, not proof.

Design research isn’t scientific. We still need to use our own interpretations of data to get to something that’s actionable. It’s therefore okay if insights turn out to be wrong. We only learn how true our insight statements are by doing more research. Even when we’re not right we’re still making informed design decisions.

This is also how we deal with sample size and confidence in our research. Insights are not a full representation of all customers but they help us make design decisions that can benefit everyone.

When working with large data sets teams can suffer from analysis paralysis. Problems can be hard to see in complex data. Insights can help cut through and give focus to a team.

How to get to insights

The data from user research will always be messy and needs analysis. It’s hard to get to good insights so research analysis is very important.

A good approach is to affinity sort observations from your research into common themes. See Natalie’s blog post about capturing user feedback to get started.

Insights should be written down as part of this research analysis. You’re looking to get to a clear actionable insight for each theme or group of observations.

Some insight examples

“People don’t know that they can turn on their heating when the weather is cold”

This is an example from our Digital Academy project Cold Weather Payments. It’s a good insight that helps us set a design challenge – ­ “how can we let people know they’ve received a cold weather payment?”.

“People on low incomes don’t know whether they’re employed or self-employed”

This example is from our live Carer's Allowance digital service. Again, it’s a good insight. It’s okay that it’s framed as a statement of fact ­– this makes it provocative and actionable.

The reality is some people understood if they were self-employed but stating this as a truth makes it easier to set a design challenge that might benefit everyone. For example “how can we capture employment information without people having to self-­determine if they’re employed or self-employed?”

Note -­ neither of these examples suggest a solution.

Making insights part of your workflow

Look for patterns in the insights you find as you iterate a product through different rounds of user research.

Eventually you should find that insights are an important part of your workflow. Your team will rely on them in their decision making and planning sessions.

They can also be a useful focal point for show & tell sessions with your team and wider stakeholders ­– a great way to communicate what you’re learning.

Leave a comment

We only ask for your email address so we know you're a real person