Fix the constraint in live work. Then hand it back.
The Implementation Sprint changes what the diagnosis found is actually preventing performance: process, standard work, management routines, accountability, measures. Technology comes in only where the diagnosis requires it. Your team owns the result.
We do not sprint to discover the problem. If the constraint is still a debate, we diagnose first.
Thirty days on one constraint.
- Redesign the stuck workflow
- Write standard work
- Install the management cadence
- Name the owner and the measure
- Enable technology only if the diagnosis funded it
- Prove it on a defined slice of live volume
- Leave a runbook
Out of scope: a new philosophy, an enterprise rebuild, vendor theater, running undesigned work, and any selling, coverage, clinical, or payment authority.
A moved constraint, or an honest account of why it did not move.
- The constraint moved, or an explicit account of why it did not
- The workflow run in live work
- Standard work and a runbook
- A scorecard tied to the scoping outcome
- A Sustain / stop decision. The default is stop.
Design in the work. Prove on a slice. Stabilize. Transfer.
Design in the work
Redesign the constraint alongside the people who run it.
Prove on a slice
Shadow Mode when useful.
Stabilize
The owner runs the huddle.
Transfer and decision
Runbook handed over. Sustain or stop.
Prove it on a defined slice while the current path still owns the customer.
Authority expands only on evidence.
One first process. Not the whole enterprise.
Insurance
FNOL completeness and routing, or submission assembly.
Revenue cycle
Denial queue standard.
Finance
Invoice-to-cash or A/P exceptions.
Customer operations
Intake classification.
Revenue growth
Pipeline hygiene and forecast cadence.
When a sprint is the right next step.
The diagnosis does not pre-commit a sprint.
Right when
The constraint and the dollars are named, an owner will take it, and a slice can be isolated.
Wrong when
The cause is still disputed, the ask is “implement the tool,” or you want an enterprise methodology in 30 days.
Not sure the constraint is proven yet?
Twenty minutes. No deck. No pitch.