The method
Systems and Empathy
Every complex human system that works is a blend of two things. The recipe is constant. The ingredients differ every time.
The diagnosis
Every organisation that is stuck sits on one side of the line.
Systems are the structure you can design. Empathy is how people actually behave inside it.
Systems-rich, empathy-poor
Tools bought, data collected, platform deployed, process documented. Nobody uses any of it as designed.
This is the AI pilot graveyard, and most large corporates.
Empathy-rich, systems-poor
Mission, conviction, relationships, extraordinary people. Nothing compounds.
This is most founder-led organisations and most foundations.
A national foundation we worked with had strong programmes and board-level ambition, and fundraising that would not move. The diagnosis found zero complete value chains and 96.6% of intended outcomes unfunded. The programmes were fine. The organisation was structurally incapable of converting them into anything.
The two conditions look identical from the inside. They need opposite fixes.
Systems
The designable part. Structure, process, governance, funding flows, decision rights. The architecture that determines what is possible.
Systems respond to intervention. They are what most consultancies sell and what software increasingly sells, because they are the tractable half.
Failure mode: a system that is internally coherent and humanly unusable.
Empathy
How people actually behave inside the structure. What they trust, avoid, work around and quietly refuse to do.
Not sentiment. Accuracy about behaviour, and you get it by asking and then believing the answer. Surface empathy can be designed. Deep empathy, the culture of a place, can only be enabled.
Failure mode: an organisation full of goodwill where nothing compounds.
Neither half is the answer. Transformation is the work of correcting the blend.
Where it came from
In May 2019 EY asked us a question about a district being built from an empty site. What actually makes a city iconic, and can it be designed on purpose?
Rather than theorise, we ran it as a live session in London, a city that had already run the experiment over four decades. We convened world-renowned authorities on cities, design, governance and public space, and let them argue while analysts modelled in parallel.
The answer that came out was not about cities.
Every complex human system that works is a blend of systems and empathy.
The principles
Guardrails, not dictate
Governance should let top-down planning and bottom-up creativity share vulnerability and evolve together. Control the edges, leave the middle alone.
Do not try to complete the design
The most important instruction we ever gave a client. A finished design cannot adapt to the people who arrive in it.
Design for future usage, not present need
Build the systems people will need based on how they imagine using them, not on how today's users cope with today's constraints.
The freedom test
A city has the balance right when women, girls and children feel free simply to be. The organisational version: can the people furthest from power act without permission?
How it resolves
The diagnosis tells you which half. The pilot proves it.
You cannot argue an organisation into knowing its own blend. You have to show it, at contained scale, in a way that is cheap to be wrong about.
At Primo Waters the pilot was one delivery route. At that foundation it was three strategies run entirely in a model. At Ogilvy it was a unit with a £10,000 flat fee and a thirty-day deadline.
We pilot so that failure is cheap. Not every pilot scales. That is the point.