Service · O-01
Readiness Assessment
A fixed-scope engagement for leaders deciding whether on-premise AI makes sense for their organisation, before any hardware is procured. In three to four weeks you get a reference architecture, a costed hardware path, a compliance map, and a clear go / no-go.
What you get
- Discovery interviews across your stakeholders
- A use-case matrix scored for value and feasibility
- A data inventory and sensitivity classification
- A reference architecture sketch for an on-premise build
- A hardware bill of materials with three price points
- A compliance map across GDPR, the EU AI Act, DORA, and NIS2
- A 3-year total-cost-of-ownership comparison against public-cloud LLMs
- A clear go / no-go recommendation
- A ~30-page PDF report and a 90-minute boardroom presentation
How it runs
The assessment runs over three to four weeks, with roughly ten founder days of work. There is a weekly check-in so there are no surprises, and the engagement closes with a final readout. Workshops run on-site or over video, whichever you prefer.
What we need from you
- An executive sponsor who can make the go / no-go call
- Around six stakeholder interviews (technical and business)
- A sample data classification, even if informal
- Existing IT context: network, identity, and document store
Indicative price
€12,000 – €18,000
Fixed-scope, with a time-and-materials cap at €18,000. Hardware is not procured at this stage; the assessment produces the bill of materials you would budget against.
Acceptance criteria
- The readout is delivered and presented
- A signed go / no-go decision is recorded
- The hardware BoM and 3-year TCO are documented
We publish acceptance criteria up front because the engineering credibility is the point.
Questions we hear often
Do we have to buy hardware to do this?
No. The Readiness Assessment is a study, not a build. We produce a hardware bill of materials with three price points so you can budget, but nothing is procured at this stage. Procurement only happens if you decide to proceed to a build engagement.
Will this lock us into Lucrion for the implementation?
No. The deliverables (reference architecture, BoM, TCO, compliance map) are yours to keep and to take to any integrator. We build on open-source and open-weights technology specifically so there is no lock-in. We would like to build it, but you are free to do so without us.
Who owns the artefacts after the engagement?
You do. The report, the architecture sketch, the data inventory, and the TCO model are delivered to you and are yours to use however you wish.
Can the workshops be remote?
Yes. Workshops and interviews run on-site or over video, whichever you prefer. Many assessments run as a mix, with a kickoff and a final readout in person where that is practical.
How does this handle data residency?
Data residency is one of the core questions the assessment answers. We map where your data lives today, where inference would happen on-premise, and what the EU AI Act, GDPR, and (where relevant) DORA and NIS2 expect of you. The compliance map documents what an on-premise design does to support each.
What happens after the assessment if we say go?
Most organisations move to a first production system, typically the Private AI Starter, using the reference architecture and BoM from the assessment as the spec. The assessment is designed to de-risk exactly that next step.
Is the price really fixed?
Yes. The engagement is scoped at a fixed €12,000–€18,000 with a cap at €18,000. If discovery reveals the scope is larger than a readiness assessment, we tell you before any extra work. We do not let it drift.
Decide with evidence, not a sales pitch.
Three to four weeks, a fixed price, and a documented go / no-go at the end. If the answer is no, you will know why.