PD Newsletter #87: digital government is a priority, not an afterthought

👋🏽👋🏻👋🏾 Hello, welcome.

We're absolutely thrilled to see that the new UK government has made digital government a priority not an afterthought, from day one. In announcing plans to bring together the digital transformation of government under a new team in the Department for Science, Innovation and Technology (DSIT), led by Peter Kyle, they are taking a momentous step towards championing digital.

The challenges and opportunities are huge. We can’t wait to see what happens next.

With so much news and commentary in the wake of a new government, this is a bumper edition of the newsletter with a mostly UK focus. We’ll be back to offering up a more global perspective next time!

Rosemary and Rob

🚀 Forwarded this email? Sign-up. Follow us on LinkedIn

A new UK government

💭 Big names in digital government (including former GDS-ers) have been sharing their thoughts on the government’s plans for digital:

➡️ Ben Welby heralds the move as a healing of the fragmentation of the digital brief.
➡️ Sarah Drummond calls for more focus on data interoperability, less on AI.
➡️ Richard Pope voices his cautions about the wording of the DSIT press release, and also sets out a bold call for a reset in public sector design, with lessons from the approach taken at early GDS.

🤔 But what will DSIT do? William Perrin draws out some key themes in his breakdown of Peter Kyle’s speech from last month’s London Tech Week, and The Guardian suggests four ways the new government can use tech to boost Britain.

🚀 As well as committing to delivering digital transformation, the new government has made being mission-driven core to its intentions, listening to the advice of Mariana Mazzucato. To stay up to date on all things mission-driven, read UCL’s Institute for Innovation and Public Purpose’s mission-oriented industrial strategy and designing and implementing mission-oriented policies.

Ways of working

❓Why “what have we already tried?” is the most powerful product question you can ask. This piece by Pavel Samsonov shows why it's critical to ensure your research is designed so that it can rule things in or rule things out if possible. “Inconclusive" is a waste if it's only there because of poor research design.

⚠️ How to escape the Plan and Execute model which fails in the complex ecosystem of tech projects by Itamar Gilad. His observations are echoed by our PD colleague Audree Fletcher in her post about why you can’t always have predictable solutions when faced with complexity: “If you’re truly committed to realising your ambitious outcomes, you need to loosen your grip on how they’re to be delivered”.

🐢 Interesting reading on the journey of one Belgium’s leading telecommunication companies, Telenet, to rewire its operating model. We particularly like the idea of “Deliberate Slowification” - rushing at complex challenges often makes them worse, so creating the space to test, learn and build the foundations for real change is almost always so much more impactful.

👉 “You will ship your org chart”: In this excellent talk by Nan Yu, Head of Product at Linear, Yu argues why Conway's Law must be taken seriously (6:10): The structure of your team will be directly reflected in its product. Make sure you take the time to plan for that.

🍎 What can you learn from looking in people's fridges? A long read showing us the value of ethnographic research from an investor's perspective. Plus more evidence as to why gathering the right data matters: In a YouGov survey asking people for their assessment of the ‘best’ decades, a pattern appeared only when the researchers considered the age of the respondents. The conclusion? People feel nostalgia not for a specific era, but for a specific age.

State of technology (sorry, it’s all AI… 🥴)

🌟 At PD we are AI curious. Ignoring AI is likely to result in missing big opportunities, but too much Kool Aid is likely to lead to expensive mistakes, or worse. Making small bets, with a focus on testing and learning, is the approach we recommend. Put simply “we encourage our clients to spend less time debating AI, and more time experimenting with how it might solve real problems. Getting your hands dirty is by far the best way to learn where AI can help your organisation, and where it cannot.”

🧠 With all the worries about AI and privacy, we’ve been following the launch of Apple Intelligence. Matt Webb wrote up an interesting summary, with some excellent stuff about Apple’s language of "world knowledge" vs "personal intelligence", a distinction which is helpful for thinking about some of the different moving parts in AI systems. And Jamie Smith shared some sharp observations on how Apple is using its brand positioning around privacy to play the ‘trust game’, and is doing an extraordinary job. This technical piece from Apple on the security design of their AI also shows an admirable emphasis on privacy and verifiable transparency which we hope will encourage others to do the same.

🤬 Meanwhile, Ludic Mataroa gives a colourfully-written data scientist's view on why companies should stop buying into AI: “Unless you are one of a tiny handful of businesses who know exactly what they're going to use AI for, you do not need AI for anything - or rather, you do not need to do anything to reap the benefits.” Plus, you don’t need to be an AI sceptic to be genuinely alarmed by the dramatic growth of Google (and other big tech companies') carbon footprint in the Gemini era.

🤨 Research conducted by the UK’s Tony Blair Institute concluded that AI could replace more than 40% of tasks performed by public sector workers: a finding which reportedly came from… asking AI. We’re not so sure about that. Neither are 404 Media, who offer a scathing (and sadly paywalled) assessment: “Automation is a complex phenomenon – in government it involves multiple levels of administration, shared standards, changing legislation, very low acceptable cost of failure.” Rachel Coldicutt also suggested some reasons to be careful to avoid too much techno-optimism in her post: ‘I'm looking for a man in AI, trust fund, 6 foot, blue eyes’.

Digital government

☀️ Guest Editor Rob posted his reflections on leaving local government to join us at Public Digital, sharing poignant observations about the trials, rewards and purpose of working at the sharp end of public service delivery, as well as five recommendations for things which should be done differently in digital local government.

🚫 Three things we must do now to prevent patient harm from digital tech. PD’s Chris Fleming writes this urgent call to action after a recent wave of tech-related failings in the NHS. He recommends committing seriously to interoperability, placing digital at the top of the table, and recognising that “digital health is not about funding ‘IT projects’ but a fundamental rewiring of the way healthcare is delivered.” Plus, Jessica Rose Morley unpacks the implications for health data from this week's King's Speech.

💡Our colleague Ross Ferguson makes the case for open roadmaps and backlogs in the public sector. Making these documents publicly available not only signals confidence and accountability for public spending, but shows a desire to engage users and work iteratively in response to feedback.

😞 Why IT procurement fails. This paper on Canada’s federal government IT contracting draws on findings from an open access research tool to assess why Canada’s IT procurement lags behind on best practice. As well as leading to scandals like ArriveCAN, these failures risk preventing any meaningful digital reform within the Canadian government. Useful lessons for governments everywhere.

💸 Seeing the value of a test-and-learn approach to tax reform in this story of the IRS’ Direct File pilot, which tested a new digital tax filing system on a limited cohort of users across the US. This work shows that success in high-risk transformation comes from decision-making with a laser focus on “executional certainty” and particularly: “maintaining the trust of users”.

🌐 Defending the internet’s open governance. In September the UN will unveil the Global Digital Compact (GDC), an agreement that will shape the future of the internet. A number of internet luminaries have weighed in with this open letter, warning against the centralised governance mandated by the GDC which threatens the “successful multistakeholder Internet governance practice that has brought us the Internet of today”.

🙌 The impressive work of UK local government: Much respect to the creators of the Birmingham Foundry at Birmingham City Council for their ambitious efforts to champion digital ways of working in spite of extremely challenging times. Also a shout out to the group using a new approach to form a public sector data strategy in Scotland: An interesting and worthwhile experiment in multi-agency co-design, as an alternative to a data strategy being cooked up in a central government office.

Something fun

🍯 Can Winnie the Pooh teach us anything about UX Design? The answer is a resounding yes.