Berthmark — Freight Rate Intelligence for Ship Operators

Ship operators make $200,000–$400,000 freight decisions dozens of times a month week using gut feel and a single phone call to their broker. We give them an AI that reads thousands of daily market reports and tells them exactly what the market is paying before they negotiate.

Muhammad Ghazwan
University of Amsterdam
BSC Economics and Business Economics

Over het initiatief / About the initiative

In welke fase zit jouw initiatief? / In what stage is your initiative?

Ideefase

Heb je jouw initiatief al gevalideerd? / Did you validate your initiative?

No

Meer informatie over jouw initiatief / More info about your initiative

Here is what actually happens when an operator gets a cargo offer.

A broker calls. He has 75,000 tons of coal loading in South Africa, discharging in Rotterdam, laycan in two weeks, offering $11.20 per ton. The operator has a few hours to decide.

He runs a rough P&L in Excel. He calls the broker back and asks what the market is doing. The broker says, “I’ve seen a couple done around $12, market feels soft.” The operator looks at the Baltic Exchange index  it covers a broad average across vessel types and routes, nothing specific enough to actually use. He makes a call.

Three days later, two identical fixtures are published in the broker reports: $12.80 and $12.60. He left $112,000 on the table. Not from bad luck. From not knowing what the market was paying at the moment he needed to decide.

This is not unusual. It is the default experience for mid-size ship operators running 5 to 50 vessels. The information that would have told him the real market rate was already out there. Hundreds of fixture reports get published every single day across broker networks. But they are unstructured texts, scattered across sources, and there is no way a person can read and synthesise them in real time.

Large operators have analyst teams who try to do this manually. Mid-size operators have nothing.

Berthmark solves exactly this. The platform reads every fixture report published daily, parses them using AI, and builds a continuously updated database of real market rates by route, vessel type, cargo type, and time of year. When an operator gets an offer, they enter the details and immediately see where that rate sits against what the market has actually concluded in the last 30 to 60 days, not a broad index, but real comparable deals.

 

Wat is er anders/nieuw aan jouw idee/oplossing t.o.v. bestaande oplossingen? / What is different/new about your idea/solution compared to existing solutions??

The tools that exist today were not built for this problem.

Voyage management systems like Veson Nautical start at $50,000 per year and are designed for back-office operations, tracking voyage costs, managing port documentation, and running accounting. They are not market intelligence tools, and most mid-size operators cannot afford them anyway.

The Baltic Exchange publishes daily benchmark indices, but they are averages across a broad panel. They tell you roughly what a standard Capesize vessel on the main iron ore route did today. They tell you nothing about your specific vessel size, your specific port, your specific cargo, or your specific counterparty.

What nobody has built is an AI-native product that reads the actual raw fixture reports and the ground-truth transaction records published by brokers every day and makes that intelligence accessible and queryable at the moment of decision.

That is specifically what Berthmark does. And the longer it runs, the more data it accumulates, the harder it becomes for anyone else to catch up.

Wat zijn jouw volgende stappen om het verder te ontwikkelen? / What are your next steps to develop the initiative?

First thing is getting in front of operators and just watching. We want to sit with people when a cargo offer actually comes in and see who they call, what they look at, and what they wish they knew before we build anything.

Then we start small. One trade route, manually curated data, and a few operators using it on real decisions. We are not trying to automate everything from day one. We want to know if the intelligence actually changes what someone does before we spend time scaling it.

The goal is a paying customer.

Wat heb je nodig om (nog meer) impact te maken met dit initiatief? / What do you need to make (more) impact with this initiative?

Three honest things.

First, introductions. We have a starting network in the industry, but we need connections to ship operators, shipowners, and maritime investors to compress months of cold outreach into weeks.

Second, data access funding. The fixture report data from Clarksons, Baltic Exchange, and broker subscriptions costs $10,000 to $30,000 per year. That is the foundation the whole product sits on, and we want to do it properly from day one rather than on scraped sources.

Third, a technical advisor who has built an AI-native B2B SaaS before. Not to build it for us, we are building it ourselves, but to pressure-test the architecture decisions as we go