It’s quarter-to-five on a Monday morning and I’m up writing weeknotes. All thanks to the cat. He jumped up on the bed around 3.30 a.m., waking me up, and my brain started pelting me with thoughts of a service assessment from last week. Today I’m writing up the report, it’s top of mind.

More of that later though. What did I get up to last week?

  • Met with Data Services, our head of product and lead product managers to talk through ideas for a unified theory of value for Digital Service Platforms
  • Met with a performance analyst from Data Services to scope the revival of the KPI dashboards from the GOV.‌UK Design System’s performance framework
  • Ran a team retrospective, joined sprint planning
  • Joined a playback on ‘cutting the mustard’ in our browser support epic
  • Got clarity on why the directorate has a focus on growth (huge thanks to Martin and Amanda)
  • Assessed a service looking to move from private to public beta, as lead assessor
  • Helped draft a blog post on how we use GitHub Projects
  • Documented the work done on our growth strategy so far, to present to our senior management team
  • Caught up with a product manager in government looking to grow their network
  • Went to Manchester and talked on a panel about growing in product management (P.S. GDS is hiring in Manchester!)
  • Gave annual feedback to half my team-mates
  • Had the usual raft of Friday one-to-ones

The theme of the week was value: defining it, delivering and measuring it, and increasing it.

Don’t ship the org chart

The Government-as-a-Platform directorate was all about shipping products that make it easier and cheaper for teams across government to build and deliver digital services. The flagship products are GOV.‌UK Notify, GOV.‌UK Pay, and GOV.‌UK Design System. There was also GOV.‌UK Registers which didn’t succeed – I don’t think teams understood it or its purpose – and GOV.‌UK Platform-as-a-Service, which is being decommissioned.

Despite being part of a unified mission, over the years the platforms have become a bit siloed. Not terribly so, but there are opportunities for sharing approaches and concept models. Partly this is due to organisational factors but also heads-down product work during growth stages. Each platform focused on its own strategy, its own set of partner services, and its own roadmap instead of having those feed into a directorate-level, overarching strategy.

So, while each product is successful in its own right, they are each chasing the same market and competing with each other for attention. And there are more service teams with a low digital maturity out there, compared to 2015, for whom a different approach is needed.

Amanda summed it up well: “We’re shipping the org chart, which isn’t what users need.” She’s right. So right. Why consider a payments platform or a notifications platform in isolation? If you’re looking to transform a service, or build one from scratch, you might need to consider both. So what can we do as a directorate to make that easier for service teams? How can we position our products and engage with services (and their decision-makers) to maximise the impact we have on making it easier and cheaper to build digital services.

It’s something I thought back in 2021, that some Zapier-like no-code service builder might join together the different platforms. And now we’ve got GOV.‌UK Forms, that’s an even more possible prospect.

I thought a lot about Amanda’s quote all week. It’s sticker-worthy, and a sign of really good product-thinking at a leadership level. It’s a sign there’s good things to come.

Growing in product

Yasmin invited me up to Manchester to speak about growing in product at one of the AutoTrader Product Talks. We answered questions on a panel, and I mostly focused on improving your understanding of strategy (so you can learn how to say no) and reflecting regularly by writing weeknotes. A lot of people touched on how working in customer support or marketing had helped them develop a user-oriented mindset, which I concurred with.

I learned a lot at NOW TV, and even more at Porism. But the biggest jump in learning happened at GDS, working on GOV.‌UK. I never felt I was doing proper product until I got onto GOV.‌UK Pay though.

None of that compared to working at a startup, though. Or working for NHS Digital.

It’d be good if GDS offered its product people a tour of duty. 3 years working across government on different teams, experiencing the problems that service teams do. It’d certainly help the org get closer to its users.

Value, data, strategy

One thing I’ve found hard on the design system is being strategic. There are few reasons not to do something, and many reasons to do everything. That’s an absence of strategy, in my mind. So if I only manage to do one thing in this post, it’ll be to create some strategic apparatus that can give the team focus in future, so they can deliver the most value for their users.

The first thing to do is improve our data foundations. There was some good work done on a performance framework but it’s not involved in decision-making much. I need to revive the dashboards, add in product-level metrics, and make these KPIs a part of team processes.

If a user comes to the design system and can’t find an element that meets their users’ needs, they’ll either go elsewhere or build their own thing. If an element is common and exists elsewhere in government, we should be serving it – or else we’re not meeting the needs of service teams. We’re here to save them time and money by offering things they can use that are trustworthy, where the costly work of design, validation and development is mostly done.

So it’s important to know what teams need, what exists elsewhere, and where we should focus our effort to deliver the most value. We don’t have a good understanding of that yet, or of the limit on value we can produce. As I’ve said before, we should be delivering 2.4 times as many benefits as costs incurred, and with us tackling more complicated elements in the backlog, we need to establish the ceiling on the effort we put in. There is a risk that things turn into vanity projects.

Railway Alphabet

Recently I’ve been taking photos of Railway Alphabet whenever I see it out in the wild. It used to be everywhere, a part of the country’s public transport infrastructure, but after the privatisation of the rail network it has been cast aside for various commercial brands. Legibility and consistency lost so that each rail company can spend money on posters and graphics rather than running trains on time. It’s a reminder of why the GOV.‌UK brand is so good, and why its survival needs protecting.

I’m going to start a blog collecting all the photos. It’d be nice to get submissions from others too. Watch this space.


I meant to read a lot more this week but my reading routine has been disrupted. I subscribed to Readwise Reader as they came out of private beta and immediately ran into an edge-case bug. I’ve paid my subscription but my account is showing as ‘Trial ended’, which means I’m locked out of my account. They’re in beta so I’m cutting them some slack, but I haven’t had an update on when it might get addressed.