Four steps of working together

It starts with where the work is actually stuck: who uses it, what they use it for, what data exists, and what must not change. The step ends with a scope that can be written down and a clear idea of what finishing means.

Loading image

Loading image

Showing the pointsWhat this step settles0.0sShowing the points

The problem and its users: which step of today’s workflow hurts, and who runs it every day.

Data and constraints: what sources exist, what rules apply, and the limits on time and technology.

Scope — and what is deliberately out of it, so it is not relitigated later.

Acceptance criteria: what "done" means, and how it is measured.

Scope, architecture and interface are settled before the first line of code. The point is to make the expensive decisions while they are still cheap to change — on paper, not in code that already runs.

Loading image

Loading image

Showing the pointsWhat this step settles0.0sShowing the points

Architecture and data flow: where data enters, where it goes, and which copy is the source of truth.

An interface reviewed before it is built, so changes happen on the drawing, not in the code.

A plan of iterations: each one a slice that runs and can be checked.

Risks and assumptions: what is still uncertain, and how it will be verified.

Work proceeds in small iterations, each ending in something that runs so it can be checked early. Rather than waiting until the end to see a product, both sides watch the same system get closer.

Loading image

Loading image

Showing the pointsWhat this step settles0.0sShowing the points

Every iteration ends with something that runs, not a progress report.

Testing happens inside the iteration, while the context is still in the author’s head.

A change of scope updates the plan and the criteria — nothing slips quietly.

Regular demos, so direction changes while changing direction is still cheap.

Delivery comes with instructions for use, and the first stretch of operation is watched closely — once people are used to it and the data has grown, that is when a system is really tested.

Loading image

Loading image

Showing the pointsWhat this step settles0.0sShowing the points

Handover includes usage documentation and a walkthrough for the people who run it.

Close attention early on, when real data starts arriving.

Feedback from users is collected and defects are fixed in priority order.

Knowledge transfer: your team can run the system itself.