Warmeston: mapping first, building second, a week apart

Warmeston ran the practical implementation programme as two separate sessions rather than one long day. The first covered the theory and mapped candidate initiatives. The second, six days later, was for prototyping and building.

Why the gap matters

Compressing both halves into a single day sounds efficient and usually is not. Mapping produces a list that feels obvious in the room and looks different once people have gone back to their actual work for a few days. Some candidates quietly turn out to be edge cases. Others that seemed minor turn out to be the thing everyone complains about on a Tuesday.

The week between sessions is when that sorting happens, at no cost. People arrive at the second session with a shorter and more honest list than the one they left with.

What each session is for

• ✳Session one: shared vocabulary, the theory that matters, and a mapped set of candidate initiatives
• ✳Between: participants live with the list while doing their normal work
• ✳Session two: prototyping the survivors, with the trainers in the room
• ✳After: what exists is a prototype tested against real work, not a slide

It is a slower format to schedule and a faster one to get value from. If you can only run one day, run the mapping day; a prototype of the wrong thing is worse than no prototype.

What got built

The prototyping session produced working things rather than plans for things. An automated auditor that checks source documents against internal rules and flags what does not match, instead of someone reading them in sequence. A collector that watches market news and brings back what is relevant rather than everything. A report generator that assembles the recurring weekly reporting from the systems that already hold the numbers. Several more followed.

The time saved across them is substantial. More useful than the total is the pattern: every one of those three replaced a recurring manual assembly job, not a piece of judgement. Nobody automated a decision. They automated the fetching and formatting that used to happen before a decision could be made.