Be it resolved that digital government is just government.
One of the best parts of this year’s FWD50 conference on public sector innovation and technology held in Ottawa and online Nov 4-6, 2024 was an Oxford-style debate chaired by conference organizer Alistair Croll. Arguing for the motion were Hillary Hartley (US Digital Response) and Alexandra MacEachern (Public Digital); arguing against were Katy Lalonde (Supply Ontario) and The Hon. Colin Deacon, Senator (Nova Scotia).
Given FWD50 is a conference focused on the digital transformation of government, Hilary and Alex benefitted from a strong head start in attendee sentiment. Arguing against the idea that “digital government is just government” took courage, intellectual dexterity, and humour on the part of Katy Lalonde and Senator Deacon. There were as many hearty laughs as there were solid arguments throughout the debate.
A good debate allows us to examine opposing viewpoints. You walk across the floor and hear the perspective of an opposing point of view, consider why your assumptions and beliefs might need an update, or even a significant overhaul. In a time when political debate rarely rises above name-calling and personal attacks, the debaters succeeded in presenting their cases, especially given they didn’t necessarily personally believe or support some of the very positions they were debating for or against.
One of the big themes of the debate was a familiar one in digital government transformation circles: centralized vs decentralized teams. Where does the work of digital government reside?
The “for” side (Hilary and Alexandra) argued decentralized. The “against” side (Katy and Senator Deacon) argued centralized. Let’s consider some of the main points.
The benefits of centralized digital teams are real:
- They can concentrate expertise, attract and retain specialized talent, providing an example to the rest of the organization as to what good is meant to look like.
- They can provide an important alignment and standard-setting function, helping establish common standards for capabilities like design, accessibility, and development.
- They have political influence and can allocate resources accordingly. For those teams located within an executive branch or adjacent to the very top of the hierarchy, the use of political leverage is often an important way to get things done in a slow-moving bureaucracy reluctant or resistant to change.
The benefits of decentralized digital teams are equally compelling:
- They can know the problems of government closer to the citizen or policy area they’re responsible for, allowing for more tailored services that meet those specific needs.
- They can be more agile and responsive to those needs of their users though being empowered to experiment or try new things, without the politics of a centralized unit and the gravity of the top of the org chart.
- They can persist through time as the organization adapts to new ways of working, a more sustainable way of doing things, ultimately affording the scale required of all of government in tackling contemporary challenges, something a purely top-down approach will struggle with.
So which one wins?
Both, of course. And neither.
The question of where digital capabilities should live is a spatial question, asked as if placing a team of people to the left side or the right side or the top or the bottom of an org chart is going to solve our digital problems. Yes, people need to live within some form of structure … but the missing concept in the debate wasn’t a question of where, but of when.
The question of where becomes more interesting when we ask it in relation to time: “When are centralized, decentralized, or distributed capabilities most appropriate?” At what moment in the evolution of any given service or policy area do we find ourselves? And how do we best structure ourselves around that context?
In the early days of the digital government service team movement, the UK GDS, USDS, CDS, ODS, Alberta DIO, and many more were all centrally located for many of the above reasons. Ines Mergel’s research of the emergence of these teams back in 2019 describes the similarities of their approaches and structures. These organizations favoured agility and innovation, new ways of working and new tools, and positioned themselves against the stability, risk aversion, and traditional Waterfall ways often personified as “traditional IT.”
This dichotomy of agile versus waterfall or innovative versus stable can be recognized elsewhere in corporate tech thinking and frameworks, predominantly in the concept of “bimodal IT” popularized by Gartner and subsequently promoted by certain large enterprise software vendors about 10 years ago. Their concept states you have two modes of IT: Mode 1: stable and efficient; Mode 2: agile and innovative. While this exploit/explore model gets beyond the either/or debate by claiming that you need both/and, it also misses the mark in the question of when and how you get from one mode to the other.
The “missing middle,” which I believe Katy in particular was trying to articulate as the idea of a horizontal capability spanning across many verticals while arguing for the against side, is represented in Simon Wardley’s concept of trimodal IT or his recently renamed* pattern of “explorers, settlers, and town planners (EST).” While recognizing the highly problematic colonial roots of this language, I’ve yet to find better ways of thinking or talking about how groups of people can be organized across an evolving technological landscape.
The idea of the explorers, settlers, and town planners framework is to align people and teams to the stage of evolution of the technologies and capabilities they’re working with (ie. your government digital service delivery and policy context) and the corresponding characteristics you will likely find in that context.
Here’s Simon Wardley’s very useful cheat sheet for those four stages of evolution and their characteristics (commonly known as Genesis, Custom, Product/Service, and Commodity/Utility).
So while Wardley has written and tweeted lots about EST over the past decade (and you should go and read his blog posts about it here, here, and here as a start) the rough summary of the concept is as follows:
- Explorers (from Genesis to Custom Built): These are the explorers and experimenters who venture into uncharted territory, creating entirely novel concepts and technologies. They’re comfortable with high levels of uncertainty, rapid prototyping, and frequent failure. Their primary focus is on research and demonstrating what’s possible, not necessarily building production-ready solutions.
- Settlers (from Custom Built to Product/Rental Service): Settlers bridge the gap between the explorers’ experimental work and the town planners’ focus on industrialization. Their key role is to productize innovations, transforming prototypes into more robust, reliable, and marketable offerings. They excel at applied research, identifying common patterns, and refining concepts based on user feedback. Settlers aim to build trust, understanding, and in for-profit contexts, profitability.
- Town Planners Town Planners (from Product/Rental Service to Commodity/Utility Service): Town Planners are the masters of industrialization. Their expertise lies in taking proven concepts and scaling them to create highly efficient, standardized, and cost-effective services or commodity components. They focus on building platforms and infrastructure that others can build upon, driving economies of scale and delivering reliable, “good enough” solutions. Town Planners are essential for creating the stable foundation that allows explorers to innovate and settlers to productize.
Wardley uses the concept of theft as to how the three groups work together, each stealing (how about borrowing?) from each other to move good bits from the left of the map towards the right of the map and vice versa. The point he repeatedly makes is that you need all three. Each is necessary in an organization to go from idea to experiment, experiment to product/service, and finally to the scale afforded by utility and infrastructure. It’s not accomplished by taking the same team and shifting their responsibilities from explore, to expand, then to exploit as the context changes around them.
When viewed through Wardley's trimodal lens, different digital government questions emerge for us to debate further:
- How do we organize our government digital teams against this backdrop of different capabilities in their various evolutionary states and within their specific service context when there’s a strong gravitational pull towards conventional territorial descriptions and cost centres of Ministries or Departments?
- Where are the new blank spaces in the map (not the org chart), the areas for exploration for new and novel practice in digital government now that we’ve ensured some digital literacies and fluencies are more commonplace and widespread than they were 10 years ago? How do we let explorers keep exploring and recognize that as a legitimate organizational learning function instead of moving everyone towards the problems of productization and scale?
- How are we investing more in our “missing middle,” to ensure one-off things become worth re-using and diffusing further across government (e.g. funding teams responsible for common components), ensuring the progress of the past 5 to 10 years isn’t totally undone or abandoned through yet another (and inevitable) re-org?
If the FWD50 debate helped me come to clarity on anything, it was the reminder that transformation is never about simple binaries—centralized versus decentralized, innovative versus stable, agile versus waterfall. Instead, it’s about understanding the nuanced evolution of our collective capabilities through time. By considering an approach that recognizes explorers, settlers, and town planners are different and all need each other to succeed, we acknowledge that every team plays a critical role in government’s digital service delivery journey, no matter where they live.
Our challenge isn’t about choosing sides, but creating flexible organizational structures that adapt to changing citizen and societal needs while maintaining the stability of the system as a whole, bridging the product/service development gap between those two ends of the explore/exploit spectrum. Artifacts and the practice they belong to, like Wardley’s value chain mapping, afford those perceptual shifts of how we see the landscape in a new way and hopefully resolve some of these evergreen and intractable debates about the org design of government.
The future of government requires both good debates and informed action. The success of digital government transformation will be measured not by where we place our teams, but by how effectively we enable them to work together through time across this evolutionary canvas.
Many thanks once again to Alistair, Hilary, and all the other speakers and participants at FWD50 for convening and holding space for these conversations (and debates) to happen.