Loading
Replacing forty spreadsheets with one operations platform
A port services operator running its entire scheduling function on shared spreadsheets, rebuilt as a single system its planners actually wanted to use.
- Client
- Halden Maritime
- Industry
- Port logistics
- Location
- Rotterdam, Netherlands
- Year
- 2025
Services
Our role. Discovery, service definition, product design and platform delivery.
01The problem
Halden Maritime coordinates pilotage, towage and berth allocation across three terminals. When we arrived, that coordination happened in a shared drive containing forty-one spreadsheets, several of which had been maintained by the same two planners for over a decade.
The spreadsheets worked, in the sense that ships got berthed. But nobody outside the planning room could see the schedule, a change made at 06:00 was not visible to the commercial team until someone emailed it, and the operation’s institutional knowledge was held by four people, two of whom were within five years of retirement.
Two previous attempts to replace the system had been abandoned. Both had started from the software, not from the work.
02What we did
- 01
Sit in the planning room
We spent three weeks in the operations room across all shift patterns before proposing anything. Most of what the spreadsheets did turned out to be undocumented rules about tide windows, pilot availability and which berths could accept which hull.
- 02
Model the rules, not the screens
We wrote the scheduling constraints down as an explicit model and validated them with the planners line by line. Several rules contradicted each other; resolving those was the single most valuable week of the engagement.
- 03
Design for the exception
Planning is exception handling. We designed the interface around what happens when a vessel arrives late, a pilot is unavailable or weather closes a window — the ordinary case largely designs itself.
- 04
Migrate one terminal at a time
We ran the platform alongside the spreadsheets for a full quarter at the smallest terminal, then moved the other two once the planners trusted it more than the file they had written themselves.
03What we built
A single scheduling platform holding berth allocation, pilotage and towage in one timeline, with the constraint model applied as guidance rather than enforcement — planners can override any rule, and the system records why.
A read-only operational view for the commercial and customer teams, which removed most of the internal email traffic the planners were previously fielding.
An audit trail on every change, which the operation had never had and which turned out to matter more to Halden’s insurers than to its planners.
04Outcome
- Berth planning is no longer dependent on four individuals. Three new planners have been trained on the platform without shadowing a spreadsheet owner for months first.
- Schedule changes are visible to the commercial team immediately, ending a long-running dispute about who was told what and when.
- The scheduling rules now exist in writing, reviewed quarterly, rather than in the memory of the people who wrote them.
- Spreadsheets retired
- 41
- Terminals migrated
- 3
- Planner training time
- Weeks, not quarters
“The difference was that they learned the operation before they designed anything. The two attempts before this one showed us software in the first fortnight. Avenor showed us our own rules written down, and several of them were wrong.”
More work
Brightmoor Trust
A grant application that does not punish small organisations
A funder whose application process was quietly selecting for applicants with professional bid writers, redesigned around the organisations it was meant to reach.
Read the case study
Nordvik Energy
Making a wind portfolio legible to the people who finance it
An operator whose asset data was accurate, complete, and impossible for anyone outside engineering to interpret.
Read the case studyNext step
Tell us what is not working. We will tell you what we think.
A first conversation is an hour, costs nothing, and usually ends with a straight answer about whether we are the right studio for the problem.