I turn digital technology into lasting capability.

Make room for invention on purpose

Innovation requires dedicated time, space, and a mandate to experiment without permission.

Invention does not happen in the gaps between deliverables. Every quarter the department runs a hackathon: no phones, pizza, teams, and one challenge — either a hard problem inside a running project or an initiative nobody has had time to start. Two days of marathon focus. It buys three things at once: a reason to reflect, a check that the work we are doing is meaningful, and a standing invitation to try something new without asking permission first.

Give structure, purpose and direction — then get out of the way

Establish clear rules and cross-functional alliances, set strategic goals, and then let the team execute.

A team that knows the rules of the game organises itself. I wrote a playbook for the department that sets them out: how we work agile, which events matter and why, how we set strategic quarterly goals with OKRs, and how we break those into sprints small enough to finish. Crucially, I build alliances across the business so these frameworks integrate with other departments rather than creating a silo. One rule sits above the others — do what delivers value and drop the rest, even when the rest was on the plan. Get the system of work right and the product largely builds itself.

Technology integration is expectation management, not architecture

Successful integration relies on managing user expectations and adapting processes, not just software architecture.

It starts before a tool is chosen, with an honest make or buy decision: build it, buy it from the market, wait for the next release, or decide not to do it at all. Then comes sponsorship — framing requirements as user stories and personas, with those users active throughout the integration, because if they are not happy there is no adoption. Architecture and compliance matter, but they are rarely what sinks an integration. What sinks it is forgetting the processes that have to change and the training that has to come with them.

New technology usually uncovers a process that was already broken. The organisation then has to accept two things at once: that it does the work wrong today, and that doing it right will take extra effort. Most people agree to that readily. The hard part is that agreeing can blow up the budget and the timeline — which is a conversation to have openly at the start, not a discovery to make later.

Plan to maintain, from the day you decide to build

The true cost and value of technology lie in its long-term maintenance, not its launch.

Developers and integrators think about making technology and putting it into service. Far fewer think about handing it to someone. Maintenance is nobody's favourite part — it is not creating and it is not launching — and it is where most projects actually fail: nobody keeps the system alive, nobody trains the next set of users, nobody decides when it should be switched off. This is endemic to innovation programmes. Early in a company's digitalisation, running experiments you never intend to maintain is a reasonable trade. Once an organisation has real digital capability that trade stops paying, and looking hard at the maintenance phase and the value case before starting is what saves the entire cost of the technology.

Move technology from liability to protected value

Introduce services at low margins until their value is proven, turning IT from a pure cost centre into protected business value.

In utilities, IT budget is read as OPEX — a cost — rather than CAPEX invested in something physical. That position on the balance sheet is why technology struggles to earn the recognition, and the maintenance budget, that it needs. My approach is to introduce new technological services at a deliberately low margin until the value is visible: a new capability, a better quality of service, information the business did not have before. Once it is visible, the service becomes part of the critical value that gets protected at all costs. The cost is not infinite either — to keep development at a sustainable pace I make the organisational changes that let us secure excellent capability, nationally and internationally, at genuinely good value.

The customer of my customer is my customer

Solve the actual problem of the end user downstream, not just the perceived problem of the project sponsor.

The person sponsoring a programme is usually not the person feeling the pain. I have watched innovators lose themselves solving the problem of whoever brings the cash, only to find they built something nobody needed. It is avoidable: look past the sponsor to the people downstream, from the very beginning, with a clear interview framework rather than an assumption. My team applies this to every new contract — day to day, not as a kickoff exercise.

Meritocracy, and leading by example

Lead from the front, maintain uncompromisingly high standards, and reserve the highest praise for truly exceptional work.

I do it and show it first. Then I expect everyone to meet a better standard, every time. My part is demonstrating that it is possible, theirs is showing they can do it — and then doing what looked impossible. While I consistently offer positive feedback to foster a healthy team atmosphere, I will always address work that falls short of our established baseline. I do not intentionally hoard my praise; rather, I constantly benchmark our efforts against the latest best work we have produced. Because of this, I reserve a genuine "Wow" and a thoughtful shout-out only for those occasions when the team truly achieves the impossible.

Aim to be first — and know when not to

Recognise when to invent and when to be a fast follower, to optimise time to market and budget.

Being first is hard enough that without the intention you fall short of everything. That mindset is why we lead our field at Axpo, and why City Energy Analyst became a worldwide success rather than another research prototype. But it is not a universal rule, and I would distrust anyone who claimed it was. An organisation that can be an excellent fast follower — with no regulatory or economic pressure forcing its hand — should follow rather than reinvent. I do not work in those organisations; I need constraints more than goals to produce something of real quality. If I did, I would choose the fast follower strategy myself, and never more so than now, when data and AI have cut time to market for most digital technology by an order of magnitude.