01 · Define
Name the result before the technology.
We write down the business measure, its baseline, the people whose work changes and the constraints on data, security and procurement. The outcome brief is a page or two, and it has an owner.
Method
One standard for delivery, whatever the technology.
Level9 works on important systems where technical decisions must produce reliable business results. The same four-stage loop runs for an AI capability and for an authentication platform, and each stage ends with a decision.
Delivery control loop
Evidence gates every stage.
- Define Name the business measure, the baseline, the users and the constraints. Gate · Outcome brief
- Test Challenge the riskiest assumption with representative data and real users. Gate · Evidence plan
- Deliver Integrate the system, the controls and the operating process. Gate · System boundary
- Measure Track adoption and realised value, then improve or stop. Gate · Value review
02 · Test
Gather evidence before committing to scale.
We find the assumption most likely to be wrong and test it with the smallest credible intervention: a proof of value for AI, a parallel-running evaluation for Radiator. The evidence plan says what would make us stop.
03 · Deliver
Integrate the technology, the controls and the operating process.
The system boundary records what data flows where, who can see it and where the human decision sits. Security and architecture review run alongside the build, and the service has a named owner before it goes live.
04 · Measure
Track adoption and realised value, then improve or stop.
The value review compares the measure with its baseline using data the organisation already trusts. If the value is short of the case, the review says why and what to do about it.
Inspectable delivery
Useful work leaves useful evidence.
Every engagement produces four short documents you keep, written for the owner of the outcome and readable cold by a reviewer, an assessor or a successor.
Outcome brief
Measure, owner, users and constraints.
Evidence plan
Assumptions, evaluation and the decision gate.
System boundary
Data, identity, policy and controls.
Value review
Adoption, realised result and next action.
Principles
What stays constant.
- Senior people stay close to the work. The person who scopes the engagement is on it.
- Evidence beats momentum. A clear stop decision is a good outcome.
- Delivery includes operating change and adoption, not just a prototype or a server.
- Human accountability is explicit for every decision a system influences.
- UK contracting and support provide local accountability.
Start with the outcome