Measuring the impact of design

I’ve been talking about the measurable impact of design since my first Nationwide project: Transform Replacement.

We showed the value of our work with clear examples.

For instance, we saved around five seconds on each click path from a customer summary to a transaction. That doesn’t sound like much. But at scale, it added up to the equivalent of 25 full-time roles from one set of UI changes alone. And there were many more improvements like this.

We also reduced avoidable errors.

This was especially true when colleagues updated personal details and addresses. At the time, these errors cost Nationwide Building Society £6.6m each year. Around £4m came from fixing mistakes.

We redesigned the flows to make them easier and more forgiving. We estimated this would save about £2m a year!

Development teams also reused our designs across other colleague-facing systems. They used the same patterns and components. On a visit to the Bath branch, I saw an ID scanning tool using our designs. It was good to see our design patterns being reused (and purposefully too).

On a later open banking project with Openwrks, we focused on reducing call times in Collections and Recovery.

If members can complete more tasks themselves before a call, those calls become shorter and more useful.

I used impact mapping with the team to shape a wider programme of work. There was plenty more to do beyond the first phase.

Impact mapping gives a simple way to link ideas to real business outcomes.

It makes assumptions visible. It also helps teams explore different ways to reach a goal, not just the first idea that comes to mind.

When Openwrks moved into pilot, they needed to show value. Both real and expected. I brought senior leaders together to define a vision for self-service.

We explored what the future could look like. That meant agreeing on the values we wanted to aim for. Not just reducing call times, but improving the overall experience.

Destructive metrics

About a year earlier, I invited Cennydd Bowles to run a masterclass on practical design ethics.

He introduced the idea of mutually destructive metrics. These are pairs of measures that balance each other.

For example:
Increase the speed of an application.
But never at the cost of people understanding what they’ve signed up for.

This idea stuck with me.

I wanted to use it to shape how we measure success in financial services.
What should we optimise for when we say we are “building society, nationwide”?

Most programmes start with a financial case.
Save money. Reduce risk. Increase revenue.

That’s often where the problem begins.

If we accept those goals without question, we make a choice. We put ethics behind profit.