Problem 01
Supplier and counterpart context resets every cycle
Due diligence done at award is rarely refreshed. By year two of a program, the official picture of suppliers, owners and exposures no longer matches reality.
Public Sector & Government
Public money, private-grade intelligence — with the audit trail built in.
For central & regional administrations, agencies and public investment vehicles.
From EU-funded programmes to national procurement, public bodies decide on complex networks of beneficiaries and suppliers. Compass helps allocate, monitor and defend those decisions with an evidence-cited, sovereign-deployable platform.
The problem
Grant and procurement decisions are audited years later. Beneficiary & supplier networks hide overlapping ownership. Cross-agency data stays siloed, so the same red flag surfaces in three places and is caught in none. Every FOI or Corte dei Conti request is a manual reconstruction.
Data we integrate
- 01Grant & procurement systems, beneficiary registries, tender portals
- 02National & EU company registries, beneficial ownership, transparency portals
- 03Sanctions & PEP lists, court records, gazette publications
- 04Cross-agency datasets (as authorized) — fiscal, social, land, transport
- 05Public documents (PDF) — laws, tender specs, contracts — parsed into structured clauses
- 06Media, community & complaint feeds (multilingual)
How we contextualize
- Entity resolution across beneficiaries, suppliers and their ultimate owners — cross-agency, cross-programme.
- Integrity signals on grants and tenders — collusion, straw beneficiaries, sequential filings.
- Explainable decisions: every allocation, denial or investigation carries the evidence and rule chain.
- Sovereign deployment, role-based access, HITL controls and immutable audit trail by design.
What's broken today
The structural problems we solve.
Problem 02
Sanctions, export-control and PEP signals scattered
Lists are checked at onboarding and forgotten. Material changes — new ownership, new designations, new adverse media — reach program managers weeks or months late.
Problem 03
AI tools that can't be deployed sovereignly
Most commercial AI requires data to leave the jurisdiction. For sensitive public workloads, that is a non-starter — so teams default to manual review and spreadsheets.
AI Skills for this industry
Trade & Supply Chain
Tender Integrity Monitor
Scans public and private tenders for collusion signals, hidden beneficial owners and abnormal bidding patterns.
Run skill
Risk & Compliance
Third-Party Due Diligence
Builds an evidence-backed dossier on any counterparty — ownership, litigation, sanctions, adverse media — in minutes instead of days.
Run skill
Risk & Compliance
Sanctions Screening
Screens counterparties and payments against EU, UN, OFAC and national lists — with explainable hits and lower false-positive rates.
Run skill
Investigation
Adverse Media Sweep
Runs multilingual adverse media checks across news, court filings and gazettes — deduplicated, ranked and cited.
Run skill
Investigation
Case File Builder
Assembles a court-ready case file from fragmented evidence — timelines, actors, sources, chain of custody — with every claim cited.
Run skill
Risk & Compliance
Regulatory Change Monitor
Tracks regulatory updates across jurisdictions, maps them to impacted processes and controls, and drafts change tickets automatically.
Run skill
What you get out
Reports, decks, Excel, dashboards and alerts — ready to ship.
Brief (PDF)
Beneficiary due-diligence dossier
Per-beneficiary PDF: ownership, prior awards, litigation, sanctions/PEP hits, red flags — cited.
Graph
Beneficiary & tender network
Interactive graph of related entities across programmes and years.
Excel
Programme allocation model
Structured Excel with proposed allocations, risk scores, evidence links and decision rationale.
Dashboard
Programme oversight dashboard
Live view of programme risk: integrity flags, disbursements, red flags per region/measure.
Audit log
Court-of-audit ready log
Every query, decision, override and export logged — inspectable by internal and external auditors.
A concrete case
€1.2B recovery-fund tranche: 340 flagged beneficiaries, 42 recovered.
A national managing authority processed a €1.2B tranche. Compass cross-matched beneficiaries with company registries, prior grant history and adverse media, and produced a ranked list of 340 files with anomalies. 42 disbursements were suspended before payment, with cited dossiers filed to the Court of Auditors — no manual reconstruction required.
Order of magnitude
What public-sector teams see in the first two quarters.
100%
Decisions with evidence trail
Every recommendation carries sources, model version and reviewer — built for oversight.
weeks → days
Time to refresh counterpart context
Continuous monitoring replaces periodic, manual due-diligence cycles.
+3×
Risk signals surfaced per program
Sanctions, ownership, adverse media and operational signals correlated in one view.
EU
Data residency and operations
Sovereign deployment, EU-resident infrastructure, no cross-border data flow by default.
How Compass operates
Your weekly public-sector operating rhythm.
- 01
Program watchlist refresh
Every Monday, programs are re-ranked by counterpart movement: new ownership, new designations, new adverse media in the last 7 days.
- 02
Supplier and third-party queue
Material changes on suppliers and sub-contractors routed to the responsible program officer with a recommended action and an evidence pack.
- 03
Sanctions and export-control review
Designations, dual-use indicators and beneficial-owner shifts matched to active programs — escalated to compliance with one click.
- 04
Oversight-ready brief
One view per program — counterparts, exposures, decisions made and evidence — ready for internal control, audit and parliamentary oversight.
Outcomes
What changes when you operate with Compass.
- Continuous, auditable visibility on programs and counterparts
- Earlier detection of supplier, sanctions and ownership risk
- Human-in-the-loop recommendations with full evidence trail
- Sovereign deployment compatible with national requirements
- Continuity of institutional context across team rotations
- Defensible decisions in front of oversight and audit
Common questions
What public-sector leaders ask us first.
- Can Compass be deployed sovereignly?
- Yes. Compass supports EU-resident and sovereign deployments, including isolated environments. See our Sovereign AI approach for the full architecture.
- Is every recommendation auditable?
- Yes. Every recommendation carries the underlying signals, sources, model version and human reviewer — designed for internal control, audit and oversight.
- Does Compass replace our procurement or case-management systems?
- No. Compass enriches and connects the systems you already operate, surfacing relational and risk context the systems alone cannot show.
Trust by design
Evidence, explainability, audit — built into every output.
Source-level evidence
Every signal, score and recommendation links back to the underlying record — the email, the filing, the transaction, the document — with the source visible at a click.
Explainable recommendations
No black-box outputs. Each priority states why it was promoted, which signals contributed, which were declassed, and which model version produced it.
Audit trail by default
Signals, model versions, reviewers and actions are logged and reconstructable later — ready for internal control, regulators and your own governance.
Human-in-the-loop
Compass recommends. Authorized people decide. Every consequential action requires a named reviewer who can confirm, override or annotate the reasoning.
See Compass on your sector.
30 minutes. Your accounts, your pipeline, your signals. We'll show you the actions Compass would recommend today.