The conventional consulting model tends toward duration. Long relationships, rolling retainers, extended programmes. There are good reasons for this — context accumulates, trust develops, complex change takes time. But there is also a less flattering reason: longer engagements are more commercially comfortable for the consultant.
I have structured City Software Engineering around a different model, and I want to explain why.
What a bounded engagement forces
A fixed scope and a defined endpoint change how both sides behave. When time is constrained, the question of what actually matters becomes urgent. Conversations that might otherwise drift toward the comfortable and the peripheral get focused. Decisions that might otherwise be deferred get made.
This is not just a commercial preference. It is, I think, better for the work.
Some of the most useful things I have done for clients have taken days, not months. A structured diagnostic process that identified the actual root cause of a scaling problem that had been misdiagnosed for over a year. An architecture review that reframed a platform decision and prevented a significant investment in the wrong direction. A short engagement at the beginning of a build that produced clarity on what should actually be built — saving months of work on what would have been built otherwise.
None of those required a long relationship. They required focused attention on the right question at the right moment.
The longer engagement problem
Extended engagements accumulate their own complexity. The consultant becomes part of the organisation in ways that can erode the independent perspective that made the relationship valuable in the first place. Decisions get deferred because there is always more time. The work expands to fill the time available.
There is also a dependency problem. An organisation that has relied on external expertise for months is in a structurally different position from one that has had a focused intervention and continued under its own steam. The former has, to some degree, externalised its thinking. The latter has had its thinking sharpened.
I want to leave clients in a better position than I found them — not in an ongoing relationship with me.
What this requires from both sides
A bounded engagement is more demanding upfront. It requires a clearer articulation of the problem, a tighter definition of what success looks like, and a willingness to prioritise. Some clients find this uncomfortable — particularly those whose situation is genuinely complex and whose instinct is to explain that complexity before committing to a scope.
That instinct is understandable. It is also, usually, addressable. Most complex situations have a specific, bounded question at their core: is this system’s structure actually the problem, or is it something else? What are the two or three decisions that will most affect this platform’s trajectory? What is the minimum viable structural change that would get this system back to predictable behaviour?
Finding that question — and then answering it with focused attention — is the thing I am trying to do. The boundaries are not a constraint on the work. They are part of what makes the work good.
Working through something like this?