In February 2023, Apolitical hosted two masterclasses for UK senior civil servants enrolled in our Digital Excellence Programme, jointly developed with the Central Digital & Data Office, Government Skills & Curriculum Unit, the London School of Economics and Political Science, and EY.
We’ve summarised the key takeaways from both sessions below. It’s a rich source of insight and practical advice for senior leaders on building a stronger digital and data culture within their departments.
➡️ Building a Digital Culture in Government
📆 Tuesday 28 February
🎤 Pamela Dow, Chief Operating Office, Civic Future
Over the course of her two-decade-plus career in public policy, Pamela Dow has gained a wealth of experience in roles spanning strategy, policy, communications and delivery. Through that experience, Dow has witnessed first-hand some of the big challenges and mistakes that leaders and organisations often encounter with digital change programmes. She walked through the seven sins of digitisation:
Believing the hype: don’t get caught up with eye-catching new tech and silver-bullet solutions from external suppliers. Blaming the people and not the product: if users don’t want to use your service, there’s something wrong with the service, not the users. Not trying it yourself: however senior you are, it’s essential that you put yourself in the user’s shoes and test the services your teams build. Underinvesting in workforce training: everyone, from policy to frontline delivery, needs the time and support to learn and adapt to new systems. Scrimping: money is always tight in government but senior leaders need to be able to make a sophisticated and compelling case for why upfront investment can deliver long-term savings. Getting too fixated on behavioural insights: people and behaviours change over time, the services government builds and tools it uses need to reflect that. Not seeing the big picture: senior leaders must be mindful of the wider political and geopolitical context when making technology procurement decisions.
🎤 Alexander Evans OBE, MPA in Data Science for Public Policy Programme Director, London School of Economics
Alexander Evans has worked at the heart of government, confronting complex security and geopolitical issues with a diverse range of global stakeholders. He is trained to see the wood for the trees and in his short presentation Evans shared his perspective on what it will take to build a richer digital culture in government.
Evans started by taking stock of government culture. Civil service culture is currently shaped by the written word. For Evans, that is the starting point for thinking about what it will take to shift government from an analogue to a digital-first culture. One that runs on data in a meaningful way – not just reports and dashboards but sharp, thoughtful analysis, a strong emphasis on visualisation and application to decision-making.
Evans also made the case that a digital culture in government does not spell the end of qualitative human insight. If anything, the opposite is true. Digital tools provide enormous capacity to collect data in the trillions of bytes. But sorting all that information requires experienced, nuanced thinking that bots can’t quite replicate.
Evans closed by thinking through what this change means for Senior Civil Servant (SCS) grades. Specialised skills in data analysis and engineering will remain the domain of specialists. But SCS need to be able to govern this work effectively, in particular as intelligent commissioners and customers of data science and technology. That means working through complex issues like developing effective legislative frameworks for new technologies; channelling investment into new capabilities; and resolving thorny ethical challenges around artificial intelligence and privacy. One of the most effective ways to drive this change, Evans said, is making it clear that colleagues’ careers will prosper if they engage seriously with digital.
➡️ Building a Data Culture in Government
📆 Thursday 2 March
🎤 Kenneth Benoit, Director of the Data Science Institute, LSE
Professor Benoit has spent a lot of time thinking, researching and writing about the intersection of politics, government and data. In his talk, Benoit outlined five key challenges that leaders need to confront to build stronger digital capability.
Recognise that everything involves data science. But also recognise that data science requires more than just advanced computing and quantitative skills. Domain experts – people with first-hand experience and knowledge about a problem area – are essential to unlocking the potential of these tools.
Disciplinary silos are a barrier. Just as academia is evolving to be more interdisciplinary, so, too, must government.
Digital excellence must be developed at every level. That includes all specialisations, not just those directly working in digital, data and technology.
Data science must be understood in the context of decision-making. Leaders need to be able to understand the risks of using evidence generated by digital tools. That means being able to critically evaluate evidence, particularly in relation to false positives and false negatives.
We’re on the cusp of an AI revolution. Government leaders need to get to grips with the potentially profound implications for education, legal frameworks, and intellectual property posed by artificial intelligence tools.
🎤 Julien Cornebise, PhD, Honorary Associate Professor, UCL
As an early researcher at Google’s DeepMind, Julien Cornebise understands the potential of AI and other advanced data systems. He has worked extensively with NGOs and government agencies. In his talk, Cornebise made the case for why some of the simpler, less celebrated methods of data collection and analysis can already produce powerful results, bringing early value from a data culture.
Data doesn’t need to be big to be powerful. Tools like ChatGPT need to be trained on enormous datasets to produce reliable results. But that doesn’t mean organisations can’t still generate valuable insights from the smaller sets they have. Data doesn’t have to be quantitative to be useful. Neatly standardised spreadsheets do make life easier. But there’s also value in messier, text-based data sets, like email, calendars, reports, and contracts – obviously accounting for proper privacy and ethics considerations.
But it does need experts to unlock insights. And those experts, Cornebise said, are the people working on the frontlines of where the data is generated. Domain expertise is about deep, first-hand experience of a problem that can provide a qualitative framework to organise quantitative research and data analysis.
You can learn more about the Digital Excellence Programme by visiting this link.