You open your tracking platform and see one arrival date, but your 3PL's system shows another, and the carrier portal says something different entirely. All these ETAs lead to lower confidence about when your shipment will actually arrive.

This is more than a minor inconvenience. It’s a costly structural leak in your supply chain that costs enterprise companies millions of dollars in profit every year.

Why your tracking platforms show different ETAs

But why are ETAs so hard to trust? To start, it’s commonplace that a single ocean freight shipment can generate three to four conflicting predicted arrival dates at any given time. This happens for a few reasons:

• Data sources: Carriers, data aggregators, port community systems, and vessel intelligence providers all pull from different underlying inputs. For instance, a data aggregator compiling tracking data across multiple carrier APIs sees different information than a carrier system built on vessel schedule data.
• Algorithms: Each provider applies their own prediction model, with different weighting of historical data, route patterns, and real-time signals. There is no industry standard for how an ETA should be calculated, yet.
• Update frequencies Some providers update every few hours. Others refresh every 24 hours or more. A prediction generated from data that is 12 hours stale can look very different from one built on real-time inputs.

For an operations manager responsible for hundreds of shipments, this creates a question that, up until now, no existing tool has answered well: which ETA should I actually plan around?

What 60 per cent schedule reliability is costing you

Global ocean freight schedule reliability sits at approximately 60 per cent, according to research from Sea Intelligence. That means four out of every ten shipments do not arrive when originally predicted.

The real problem is not the volume of ETA data, it is knowing which of it to trust. Behind every missed shipment window is the same question: which of the ETAs was actually right? More data has not made that question easier to answer.

Ultimately, this uncertainty translates directly into financial losses to your bottom line. These figures reflect commonly cited industry estimates across multiple cost categories.

• Safety stock inflation: Operations teams buffer inventory as safety against unpredictable arrivals. A conservative estimate puts 10 to 20 per cent of inventory carrying costs that are directly related to unreliable ETAs.
• Demurrage and detention: When free-time windows are missed, charges accumulate. For large enterprise shippers, D&D costs can exceed tens of millions annually.
• Warehouse inefficiency: Standby labor and overtime are triggered by shipments arriving outside planned windows. These costs, which should be avoidable, scale directly with volume.
• Retail markdowns: For fashion and seasonal goods, a container arriving two weeks late can lose 30 to 50 per cent of its retail value due to missed sell-through windows.
• Emergency freight: When ocean shipments are unreliable, teams resort to air freight at five to ten times the original ocean rate.

The problem is not that disruptions happen. The problem is that teams learn about them too late to plan efficiently, and that causes costly leaks that add up to a big financial problem.

Why adding more data sources doesn't solve the problem

The industry's response to conflicting ETAs has been to subscribe to more tracking providers. But more sources and data doesn’t necessarily equal improved accuracy.

Why? Every tracking provider is still limited to their own data and algorithm. No single provider has an agnostic view of the full picture. This creates additional gaps:

• Single-source dependency: Most organizations default to one provider globally, even if that provider is consistently inaccurate on specific carriers, ports, or trade lanes.
• No cross-validation: There is no standard tool to compare one provider's ETA against another's for the same shipment, at the same point in time, to determine which is more likely to be correct.
• No dynamic selection: Even organizations subscribed to multiple providers have no way to automatically select the best source based on context: the carrier, the port pair, the current conditions.

The industry is collecting more data without first answering the fundamental question: how do you know which data to trust?

The shift from more data to better decisions

Think about how flight search engines like Kayak and Skyscanner work. They sit above multiple data sources and help you identify the best option based on your specific needs.

ETA prediction in ocean freight needs to work the same way. Not another tracking tool competing for accuracy, but an intelligence layer that evaluates the trustworthiness of existing predictions.

That is what Moddule is building with ETA IQ.

ETA IQ sits above your existing tracking providers. It ingests predictions from multiple sources, applies machine learning to score how much to trust each one based on the specific carrier, port, trade lane, and conditions, and surfaces a single confidence-weighted prediction you can plan around. As real-world outcomes are confirmed, the model improves over time.

You go from asking "what does this provider say?" to "which prediction should I actually trust for this exact shipment right now?"

The intelligence layer, in action

Moddule announced ETA IQ, a predictive milestone intelligence engine, at Manifest Vegas 2026. We are now in the process of building a proof of concept with a major international retailer moving more than 100,000 containers a year. The pilot is designed to validate ETA IQ's accuracy against real-world operational data at scale, comparing Moddule's predictions against the tracking providers the business already uses.

If your enterprise team is managing multiple ETAs with no reliable way to know which one to act on, book a demo to learn how ETA IQ can work for you.

‍