Lumen & Lever advises and assesses. It does not implement.
Most organisations do not need an AI tool recommendation first. They need to know what AI is already touching, whether their documents and data can support reliable use, what an implementer must satisfy, and whether the organisation is structurally ready to scale. The correct entry point depends on which question is live.
Four questions. One controlled path. Where are we now? Can our documents support AI reliably? What must a builder satisfy? Are we ready to scale?
AUD 7,500 · 1 week
AUD 15,000 · 2 weeks
Best for: Organisations where AI is already being used across teams and leadership lacks a reliable management view.
Outcome: The Baseline establishes where AI is being used, what information may be exposed, which uses can proceed, which require conditions, and which should stop. It produces an AI use register, exposure classification, proceed / conditions / stop control model, practical staff-use guidance, and a leadership briefing.
Where a workflow emerges, the report names the right next step or recommends no further action. The Baseline does not specify or build the workflow. Any implementation brief is a separate engagement.
Choose this when: Choose this when the live question is: where is AI already being used, what data may be at risk, what should staff be allowed to do, and which workflows are worth examining first?
Do not choose this when: Do not choose this when the immediate problem is technical document extraction quality, table preservation, source traceability, or production-scale capital governance.
AUD 9,500 to 12,500 · 1 week
Best for: Organisations using documents inside AI retrieval, internal search, compliance review, legal discovery, financial tracing, or document-heavy operational workflows.
Outcome: The review tests whether the source material survives extraction with enough structure for AI to use it reliably. It assesses whether tables, clauses, source pages, annotations, metadata, layout relationships, and evidence traceability are preserved or damaged before retrieval and model reasoning begins.
Where the documents support reliable use, the review names the right next step or recommends no further action.
Choose this when: Choose this when the live question is: can our leases, contracts, statements, correspondence, policies, PDFs, RTFs, or legacy documents support reliable AI use without losing meaning or traceability?
Do not choose this when: Do not choose this when the organisation still lacks visibility over general staff AI usage. Start with the Baseline. Do not choose this when the documents are already understood and the next problem is briefing the builder. Use the Implementation Control Brief.
Pricing detail: AUD 9,500 for a constrained review of one document family. AUD 12,500 where there are multiple document types, higher traceability needs, or stronger privacy and local-processing concerns.
AUD 12,500 to 18,000 · 1 to 2 weeks
Best for: Organisations that have identified a specific AI workflow worth pursuing and need a controlled brief for whoever will build or configure it.
Most AI implementation failures begin before build. A loose brief produces a loose system. The Implementation Control Brief gives the business a document the implementer can build against and the buyer can hold them to.
Outcome: The brief defines the workflow boundary, business owner, source documents or systems, data sensitivity, required outputs, human review points, evaluation criteria, acceptance thresholds, audit requirements, security constraints, exception handling, handover expectations, and production-readiness gates.
Where the brief is accepted, the next step is implementation by the client's chosen team. Lumen & Lever does not deliver implementation.
Choose this when: Choose this when the live question is: we know the workflow we want to pursue. What exactly must our internal team, current IT provider, software vendor, or specialist AI implementation partner build, prove, log, review, and hand over?
Do not choose this when: Do not choose this if the organisation still does not know which workflow matters. Start with the Baseline. If document reliability is uncertain, resolve that first with a Document Structure Review.
Pricing detail: AUD 12,500 for one clearly bounded workflow with known systems and accessible source material. AUD 18,000 where the workflow involves multiple teams, sensitive data, multiple document types, or more detailed implementer acceptance criteria.
AUD 40,000 to 55,000 · 4 weeks
Best for: Organisations preparing for broader AI scale, further capital allocation, or board-level governance decisions.
Outcome: The Sprint assesses whether the organisation has the structure required to absorb AI safely at production scale. It covers AI system inventory, governance maturity, lifecycle control, vendor exposure, cost behaviour, override and containment, data integrity, and capital gate readiness.
Choose this when: Choose this when the live question is: are we structurally ready to scale AI, release capital, govern vendors, control lifecycle risk, model production cost, and brief the board?
Do not choose this when: Do not choose this for basic staff-use visibility. Start with the Baseline. Do not choose this for a narrow document-pipeline concern. Use the Document Structure Review. Do not choose this where a single workflow has already been selected and the immediate need is an implementer-ready brief. Use the Implementation Control Brief.
AUD 8,000 to 12,000 per quarter. Banded by system count and review depth.
Once an AI workflow is live, the question changes. The issue is no longer whether implementation can proceed. The issue is whether the system remains within the conditions under which it was approved.
The Quarterly Control Review refreshes the AI use register, reviews control drift, checks whether vendor or tool changes affect the original risk position, reviews workflow performance against the agreed criteria, and updates leadership on whether the system should proceed, proceed with conditions, be remediated, or be withdrawn.
Each engagement is preceded by a neutral intake conversation that confirms the correct entry point. The intake is a routing step. If none of the interventions fits, the correct recommendation may be no engagement.
Lumen & Lever also licences Sourcetrace, a separate commercial instrument. See Sourcetrace for details.
Start the intake