"It's for everyone" is the most expensive design decision you can make. Not because it's wrong to serve everyone. Because trying to serve everyone equally means you serve nobody particularly well. 🧵

Your lagging indicators confirm what happened. Your leading indicators signal what's coming. Most organisations only track lagging indicators. So they're always reacting.

Little's Law in plain English: Average completion time = work in progress divided by throughput rate. It sounds abstract. It explains why backlogs grow even when teams work harder.

The difference between good service design and expensive guessing is evidence. Not lots of it. The right kind. Here's what that looks like in practice. 🧵

Local optimisation: improving one part of a system without considering the whole. It often creates new problems elsewhere. Here's how to spot it.

Škoda and University of Salford researchers found a frequency (aka 'safety gap') that noise-cancelling headphone algorithms can't filter out. The result is DuoBell: a bicycle bell that actually gets through. Nice bit of design thinking. www.skoda-storyboard.com/en/skoda-wor...

Why do some services become irreplaceable? It's not usually the design. It's the moat around it. Moats aren't just for businesses. They exist in public services too.

Starting with the solution isn't being proactive. It's guessing. And you might guess right. But you probably won't. And you won't know until much later, when it's expensive to change. 🧵

"Systems produce the results they're designed to produce." That sounds cynical. It isn't. It's one of the most useful ideas I've encountered in professional life. Here's why:

Strategy isn't a document. It's a set of choices about what you will and won't do. Most strategies describe goals. The hard ones describe trade-offs. That's the difference.

Today I'm launching Design the service: a free 14-day email course on service design. A thread on what makes it different, and why I built it. 🧵

One of the most useful ideas in service design: every service has a frontstage and a backstage. The frontstage is what users experience. The backstage is what makes it possible. They're not separate problems. They're two sides of the same thing. 🧵

Two things underlie most service failures: bottlenecks and incentives. Understanding both changes how you design and argue for your work.

The most valuable part of a systems map isn't the arrows. It's the assumptions you label.

Who decides what's in scope? This is a political question as much as a technical one.

Does the right information reach the right person at the right time? This simple question reveals a lot about why systems behave the way they do.

Network effects: the value of a service increases as more people use it. Below a critical threshold, adoption is hard. Above it, momentum is self-sustaining.

A jet engine is complicated. A healthcare system is complex. The difference matters more than you'd think.

Pricing is a design decision. What you charge for shapes what gets used. That makes it as much a design question as a financial one.

What could you test this week that would tell you something useful in two weeks? That's the question that unlocks change in complex systems.

Value proposition is one question. What job does someone need done? Not "what does the service do?" but "what does the person achieve by using it?"

Every system has a constraint. It's the step that limits how fast work can flow through. Find it, and you find where to focus.

A system isn't a thing. It's a pattern of behaviour. It emerges from how parts connect and interact. Change the connections, change the behaviour.

A systems map isn't a diagram of what exists. It's a diagram of what you think exists. That's what makes it useful.

I've just launched a free 7-day course on systems thinking. It's for people who keep running into problems that resist conventional solutions. Thread on what it covers:

I’ve been following DOGE with a real sense of shock from the UK. This is a good article - how understanding the relationship between policy, modern services and technology is crucial, and what is technology a means to? www.techpolicy.press/doge-underst...

✏️ ✏️ I wrote a quick little rant about the awful UX of login codes.

I’m reading ”The Great Mental Models” by @shaneaparrish.bsky.social. It’s 4 books packed with tools for better thinking: 1. General thinking concepts 2. Physics, chemistry, and biology 3. Systems and mathematics 4. Economics and human behaviour I’m two books into the series, and I highly recommend

ROI isn’t the best way to talk about the value of UX, but behaviour change is. When it comes to UX, ROI’s limitations become apparent. It’s a financial measure, while UX is about human experience. Behaviour change is a more meaningful indicator.

I fed NotebookLM all of the service assessment reports (www.gov.uk/service-stan...) from the past 18 months just to see. Got 17 minutes and wanna hear some AI-generated Americans talk about child benefit, spaceships and legacy systems? I've uploaded it here soundcloud.com/anna-goss-48...

How has Microsoft convinced so many people to use their terrible software? It’s a masterclass of con-artistry!

I miss This Is My Jam. The simplicity of it was wonderful. Anyway, this is my jam; the saxophone is awesome: open.spotify.com/track/5EGAY1...

When is an interaction designer, not an interaction designer? When they only design in Figma! 🙈🙉

Rescued by a bot, by Christoph Niemann for the New Yorker, imagines the unintended consequences of turning to artificial intelligence to solve problems of the artistic imagination. #AI