If you want to know where you stand
The ladder is the same instrument a Diagnose uses, and you can place yourself on it without speaking to anyone.
Place yourself on the ladder →The first engagement
Diagnose
Different definitions. A written register.
The first engagement is a diagnosis, not a transformation. It is short, the fee is fixed and agreed before it starts, and it ends on a defined date with a document you keep. Nothing about it obliges you to continue, and it is designed so that stopping afterwards still leaves you better off than starting.
Fixed feeDefined endA document you keep
What it is
Most companies run on definitions nobody has written down. Two departments each know what "qualified" means, and the two meanings are not the same. That gap is invisible on an org chart and expensive everywhere else: it is where reports diverge, where automation quietly produces the wrong answer, and where a project that looked finished turns out not to be.
Diagnose finds those gaps and writes them down. It is a bounded piece of work with a defined artifact and a defined end — not the first instalment of something open-ended.
What you receive
One document, in your own words, containing five things:
The conflict register.
Three to five places where two parts of the business already disagree about a definition, recorded verbatim and attributed to roles rather than to people.
The profile.
Where each assessed area sits on the adoption ladder — with anything we did not assess marked as not assessed, rather than guessed.
The binding constraint.
Which weak area is capping a strong one. A company's effective level in any area is limited by the weakest area it depends on, and naming that is usually the reason a previous attempt did not hold.
The sequence.
What has to be true before the next rung, in order.
A baseline.
Two or three numbers you already have, recorded at the start. Without a baseline captured before anything changes, no later improvement can be shown to have worked.
The register is the product. It is yours to keep, and it stays useful whether or not you ever work with us again.
Every business runs on these twelve areas
Illustrative — sample values
Each row is scored 1 to 7. The solid bar is how far that area actually works today. The dashed bar is how far the company believes it works. The gap between them is the risk.
Actually works todayBelieved to work — not earnedNot attempted
Strategy & market intelligence
Works at 3 · ConnectedBelieved 4 · AutomatedGap: small
Products & services
Works at 4 · AutomatedBelieved 5 · AI-assistedGap: small
Marketing & demand generation
Works at 3 · ConnectedBelieved 5 · AI-assistedGap: wide
Sales & CRM
Works at 2 · DigitisedBelieved 5 · AI-assistedGap: severe
Operations & service delivery
Works at 4 · AutomatedBelieved 5 · AI-assistedGap: small
Supply chain & inventory
Works at 3 · ConnectedBelieved 4 · AutomatedGap: small
Customer service & success
Works at 5 · AI-assistedBelieved 6 · Agent-enabledGap: small
Finance & administration
Works at 5 · AI-assistedBelieved 5 · AI-assistedGap: none
People & organisation
Works at 3 · ConnectedBelieved 4 · AutomatedGap: small
Knowledge & information
Works at 3 · ConnectedBelieved 5 · AI-assistedGap: wide
Technology, data, security, integrations
Works at 2 · DigitisedBelieved 4 · AutomatedGap: wide
Risk, compliance & resilience
Works at 2 · DigitisedBelieved 3 · ConnectedGap: small
What the seven levels mean
- 1Fragmented
- 2Digitised
- 3Connected
- 4Automated
- 5AI-assisted
- 6Agent-enabled
- 7Continuously optimised
How it works
It runs as a structured interview, and it works best with two people in the room who own adjacent parts of the same process. Asking each of them the same question and letting them answer in front of each other is what surfaces a disagreement that neither knew existed. That is the instrument, and it is why the engagement is short.
Before the conversation, it helps to look at the adoption ladder — seven levels, from fragmented to continuously optimised, with the gate that earns each one. Different departments will sit on different rungs. That is normal, and it is the useful finding.
What it costs
A fixed fee, agreed in writing before work starts. No subscription and no retainer. The number depends on scope — how many departments and systems are involved — and you have it before you commit, never mid-engagement.
Cost, risk and ownership covers this in full, including who owns the result after we leave and the four conditions under which we decline the work.
What it is not
It is not a software implementation, and it is not a recommendation to replace tools that already work. A new tool enters only where there is none. We define how the work runs, build the connections into systems already in use, and hand everything over — we do not stay on the pager afterwards.
It is also not a scoring exercise. A single readiness number would be easy to publish and would not survive contact with a real business, because the interesting finding is never an average — it is which one area is holding the others down.
What comes after
If the register proves useful, the next phase defines the terms it exposed precisely enough for AI to act on them, and the phase after that builds the loop that keeps those definitions true as the business changes. Each is a separate decision, quoted separately, and none is assumed by this one.
Scope tends to widen during a first conversation. The answer is always that the wider piece is the next phase, and that it will be faster because the foundation exists.
If your category is regulated
Where what gets published carries regulatory exposure — health products, medical nutrition, supplements, finance, legal — the sequencing changes. Work that ships to the public needs a named review gate with a logged approver in front of it, which is a control rather than a quality check.
Most of the value sits below that line anyway: duplicate records, reports rebuilt by hand every month, information nobody can find, data re-typed between systems. None of that is claims-exposed, and all of it is expensive. The constraint is a reason to sequence the work correctly, not a reason to postpone it.
Which rules apply depends on the jurisdiction and should be confirmed with your own advisors. What we bring is the design consequence, which holds either way.