Skip to content

Home/Sectors/Commodity trading

AIS data for commodity trading that shows where every position came from.

Build flow, balances and freight models on raw positions from our own network rather than on a finished cargo number. Delivered in real time, with receiver metadata and a trust score attached to every message.

100,000+

Vessels tracked / day

49M+

AIS positions / day

40+

Countries covered

<5s

Antenna to your API

Most AIS data for commodity trading arrives already interpreted.

The desk receives the conclusion rather than the input: sufficient for a screen, insufficient for a position.

  • The underlying messages are no longer recoverableA cargo volume, a laden or ballast leg, a floating storage figure or a transit count through Hormuz or Bab el-Mandeb arrives as a finished number. The analyst cannot distinguish a genuine draft change from a poor interpolation, or attribute the position to a receiver.

  • The input itself is now under attackSpoofing and GNSS jamming distort precisely the waters that move the market, and a dark ship-to-ship transfer presents identically to a coverage gap. A second feed can also mean the same positions purchased twice: a run of acquisitions between 2023 and 2025 brought several of the best known tracking brands under a single owner.

  • We supply the source, not a viewWe operate our own physical infrastructure, terrestrial and ship-borne, with satellite feeds for open ocean. Every message is encrypted on the hardware at capture, nanosecond time-stamped and given a trust score by our anomaly detection models. Interpretation remains with the desk.

Run us alongside your incumbent feed.Send an MMSI or IMO number. We will return what our network collects on that vessel, with receiver metadata and trust scores attached.

Benchmark against your feed

From an in-house flow model to pre-trade screening with an evidence trail.

  • Raw positions for your own flow and balances models

    Take the message stream rather than a packaged analytic and construct crude, clean and dirty products, LPG and LNG balances from inputs you can inspect. Tonne-miles, laden and ballast legs, port call sequences and draft analysis remain yours to define, and the assumptions remain inside the desk rather than a vendor model.

  • Floating storage and chokepoint counts you can audit

    A floating storage estimate or a transit count through Hormuz, Malacca, Suez or Panama is only as sound as the positions beneath it. Trust scores and receiver metadata allow a researcher to corroborate a cluster of stationary tankers before it becomes a number in a balance or a nowcast.

  • A second source that is structurally independent

    Upstream AIS supply has consolidated sharply. A run of acquisitions between 2023 and 2025 placed several of the best known tracking brands under a single owner, so nominal redundancy can resolve to the same data arriving twice. We are independently owned and operate our own network, which gives a data management function a warm second feed to validate against and a defensible position at renewal.

  • Pre-fixture and pre-trade screening with an evidence trail

    Screen a counterparty vessel before a fixture or a cargo purchase for AIS gaps, ship-to-ship transfers, shadow fleet association, flag and MMSI changes and price cap exposure. Because provenance travels with each message, compliance can evidence what was screened, on which positions, and that the record has not been altered.

  • Positions as they were received, not a later revision

    A backtest is tradable only if the archive is the record that existed in real time. Every position we deliver is nanosecond time-stamped at capture and encrypted on the hardware, so what reaches the desk is the message as it arrived rather than a cleaned or interpolated version of it.

  • Delivery into the stack the desk already runs

    Raw NMEA over TCP delivers the AIVDM sentence untouched for engineering ingest, a REST API serves research notebooks, and a WebSocket stream carries anything that has to move intraday, in under five seconds from antenna to your API.

How the data is produced,
and how you can check it.

Every station runs the same standardised hardware and encrypts each message at the point of capture.

Monitor

Cover the traffic you model

Track 100,000+ vessels a day across 40+ countries, from tankers and gas carriers to Capesize and Panamax dry bulk, on our own terrestrial and ship-borne receivers, plus satellite feeds for open ocean.

Verify

Know which receiver collected it

Every position is nanosecond time-stamped and given a trust score by our anomaly detection models, so a spoofed or manipulated track is flagged rather than quietly merged into a clean line on a chart.

Act

Into the model, the screen or the audit file

Take it as raw NMEA over TCP, a REST API or a WebSocket stream, in under five seconds from antenna to your API, and keep the underlying records behind whatever the desk decided to do with them.

AIS data for commodity trading: the questions desks ask.

Can AIS data be faked, and how would a trading desk know?

Yes. A transponder can broadcast a false position, and area-wide GNSS interference can distort genuine ones. The two present different patterns, and distinguishing them requires more than one track: vessel-originated spoofing tends to affect a single transponder while surrounding traffic behaves normally, whereas jamming tends to distort many vessels in the same area at once. Comparing what independent receivers collected across that window is what makes the distinction visible, which is why our positions carry receiver metadata and a trust score rather than arriving pre-smoothed.

What is the difference between raw AIS data and cargo tracking analytics?

Analytics are an interpretation layer. A third party has already determined what a draft change means, when a port call began and how much cargo moved, and the desk receives that conclusion. Raw AIS is the input to the decision: the NMEA sentences, the MMSI, the timestamps and the positions. A desk that employs quants and seeks a differentiated signal should own the interpretation rather than purchase it pre-made alongside the rest of the market.

Are you genuinely independent of the existing AIS providers?

We are. Worldwide AIS Network is independently owned and operates its own physical receivers rather than reselling another network, and we are financed by an entrepreneurial loan from Den Danske Maritime Fond, established under the Startup Denmark scheme, and building on peaq. The distinction is commercial as well as technical: a second feed provides redundancy, and restores price tension at renewal, only where it does not share an owner with the first.

Do you cover the chokepoints and load ports we model, and how do we check?

It varies by region, because we supply only what our own network collected. Specify the areas that matter to the desk and we will state plainly where our reception is strong and where it is thin, on the raw result rather than a coverage map. Every message is encrypted on the hardware at the moment of capture and nanosecond time-stamped, so what the desk assesses is the position as collected rather than a smoothed version of it.

What does an AIS gap actually prove for sanctions screening?

Less on its own than most screening workflows assume. An extended gap is treated as a red flag under OFAC guidance, but a switched-off transponder and a hole in a provider’s own coverage present identically on a single feed, so analyst time goes to false positives. Establishing whether coverage existed during the window, and which receivers were collecting at the time, is what converts a gap into either an escalation or a dismissal defensible to an auditor.

Test us against the feed you already pay for.

Send an MMSI or a chokepoint. We will return what our network collects there, and where every message originated.

Benchmark against your feed