Step 1 — Establish canonical identities
Ask three departments in a shipping company how many days a vessel was off-hire last year and you will get three answers. Not because anyone is wrong, but because operations, chartering and finance each hold a defensible definition and a separate record.
That is the problem a single source of truth solves. It is not a technology project. It is a governance project with a technology component.
Five identities carry almost all fleet data. Each needs one authoritative record, one owner, and a rule for resolving aliases.
| Identity | Key | Authority | Common failure |
|---|---|---|---|
| Vessel | IMO number | Fleet register | Name/flag change creates a duplicate |
| Equipment | Fleet-wide code | Technical | Each ship codes its own |
| Seafarer | Verified seafarer ID | Crewing | Agency feed creates a second person |
| Voyage | Voyage number + vessel | Operations | Commercial and technical number differently |
| Counterparty | Registered entity | Finance/compliance | Trading names proliferate |
If those five are right, most reporting problems disappear. If any one is wrong, no amount of dashboarding will fix it.
Step 2 — Define the terms that get argued about
Write them down, publish them, and refuse to accept a report that uses a different one.
| Term | Definition to adopt |
|---|---|
| Off-hire | Hours the vessel was unavailable under the charterparty, by cause code |
| Available days | Calendar days less off-hire, less scheduled docking |
| OPEX | Costs to operate the vessel, excluding capital, financing and voyage costs |
| Voyage costs | Bunkers, port charges, canal dues, agency, emissions allowances |
| Officer retention | Officers returning ÷ officers eligible to return |
| Critical spare | Spare whose absence prevents a critical job or affects a critical system |
| Near miss | Unplanned event that could have caused harm or loss but did not |
Seven definitions. Most of the reconciliation work in a shipping company is caused by not having them.
Step 3 — Measure data quality explicitly
Publish the bars monthly. Data quality that is not measured does not improve, and a scorecard converts an abstract complaint into a specific, assignable task.
Four dimensions worth measuring:
- Completeness — required fields populated
- Validity — values conform to the allowed set
- Uniqueness — no duplicate identities
- Timeliness — records created within an agreed window of the event
Step 4 — Sequence the work over twelve months
| Months | Work |
|---|---|
| 1–2 | Name owners for the five identities; agree the seven definitions |
| 2–4 | Build the fleet register; resolve vessel duplicates and aliases |
| 3–6 | Standardise equipment coding on three pilot vessels |
| 5–8 | De-duplicate seafarer records; structure certificate data |
| 6–9 | Unify voyage identity across commercial, technical and finance |
| 8–11 | Rationalise supplier and counterparty master |
| 10–12 | Publish quality scorecard; automate reconciliation reports |
Resist the temptation to start with the dashboard. A dashboard over bad identities produces confident, wrong answers faster than a spreadsheet does.
Step 5 — Make compliance the forcing function
If the governance argument does not land, use the regulatory one. From 2026, EU ETS covers 100% of applicable emissions on in-scope voyages and includes methane and nitrous oxide alongside CO₂. FuelEU Maritime requires GHG intensity to be calculated against a 91.16 gCO₂e/MJ baseline with compliance balances that can be banked, borrowed or pooled.
Both require the same thing: fuel consumption, by type, correctly attributed to a correctly classified voyage. That is a data model requirement dressed as an environmental regulation, and it has a cash penalty attached. Fleets that cannot link fuel to voyage cleanly will pay for the gap.
What "good" looks like at the end
- One number for off-hire, agreed by operations, chartering and finance
- A monthly owner's report generated from live systems, not assembled
- Certificate expiries that cannot be missed because assignment is blocked
- Emissions cost attributable to individual voyages within days of completion
- A supplier price history that survives a change of purchasing manager
- New vessels onboarded to the standard in weeks, not quarters
regulatory references to Directive (EU) 2023/959 and Regulation (EU) 2023/1805. The quality scorecard is an illustrative model. Reviewed by the Zeaclub Editorial Team, 24 August 2026.
Frequently asked questions
Is this a data warehouse project?
A warehouse may be part of it, but the value is in the identities and definitions. A warehouse built over inconsistent identities is an expensive way to publish disagreement.
How long does it take?
Twelve months for a mid-size fleet to reach a trustworthy baseline, with visible improvement from month three if the identity work starts immediately.
Who should own it?
Someone with authority across operations, technical, crewing and finance. In practice that means a COO or a head of fleet performance — not IT alone.
What is the first thing to fix?
The vessel register. It is the smallest dataset, the easiest win, and every other identity hangs off it.