A few years ago, I found myself questioning whether the way we defined “good” in user experience still held up. At the time, I was leading design work inside a major UK Government department, building on the foundation laid by the Government Digital Service (GDS). The GDS standard was strong but I knew we could go further.

So I extended it.

I wanted a standard that didn’t just tick boxes but actually reflected what people needed from modern services. One that accounted for complexity, life events and the reality that people don’t always engage on our terms. I built a new set of principles. More human. More complete. Still grounded in accessibility and simplicity but broadened to reflect a deeper, more responsive experience.

Designing with Purpose

At its core, good user experience should solve a whole problem. That means:

• Accessibility by default
• Trust that’s earned
• Support that’s available before someone has to ask for it

I added ideas like being reactively proactive — designing services that respond to life events before someone has to struggle through them. I put vulnerability at the centre, not off to the side as an edge case. And I created space for cross-channel journeys because real life doesn’t always happen in a single browser tab.

Modernising the Standard

Where this thinking has matured most is in how we treat data and the introduction of AI.

Data is no longer just back-office plumbing. It’s a design tool. When used properly, it reduces effort, eliminates mistakes and opens the door to tailored experiences. No one should have to tell us the same thing five times in any product or service.

AI pushes this further. It gives us the ability to:

• Predict what someone might need next
• Offer support in real time, not after the fact
• Personalise content and interaction based on real context

But that only works when done transparently and ethically. This isn’t about replacing humans. It’s about designing systems that feel more human in how they respond.

The Good Design Standard

Here’s what the standard looks like now, including what good means and how AI and data play a role where relevant:

1. Accessible

Good is: Designing every part of the journey to be usable by everyone, from notifications to forms. Accessibility is baked in, not bolted on.

2. Findable

Good is: Ensuring users can find your service easily and know what to expect. Visibility, trust and accurate signposting are key.

3. Joined up

Good is: Designing across key needs and/or policies so journeys don’t feel like a maze. Solve the whole problem, not just your bit.

4. Simple

Good is: Making things intuitive and reducing mental load. Use research to design for clarity and use AI to streamline where it helps.

5. Secure and Respect Privacy

Good is: Handling personal data responsibly. Be consistent with authentication and use AI in a way that protects trust and enhances security.

6. Supportive

Good is: Meeting users where they are, with tools and real-time help. AI can support this by offering intelligent prompts or chat-based help when appropriate.

7. Reactively Proactive

Good is: Responding to changes in someone’s life before they have to explain themselves. Use event-driven data and AI to provide relevant, timely help.

8. Meet a Need or Needs

Good is: Using research to understand what people want to achieve, even across boundaries. AI helps personalise the journey and identify emerging needs.

9. Consistent

Good is: Speaking the same language and using repeatable design patterns. AI-generated content and interaction still needs to follow these same rules.

10. Cross-Channel

Good is: Letting users start on one channel and finish on another without friction. AI can adapt content and interface to match the channel and context.

11. Use Data

Good is: Creating seamless journeys by using what we already know, responsibly. Personalise where it helps and automate the mundane to give people back time.

12. Inform and Notify

Good is: Keeping users updated using their preferred contact methods. AI can generate helpful, contextual messages but they must stay useful and human.

---

This standard started as an evolution of the GDS approach but it became something more. It’s a reflection of where we are now and where we’re going.

If you’re designing anything that matters, I hope this helps. We don’t need more complexity. We need more care, more curiosity and more commitment to doing things right.

Because good design is never just about interfaces. It’s about people. Always.