The industry doesn't need more ETAs. It needs to know which one to trust.
In logistics, a wrong ETA costs real money. Each slip equals missed seasons, halted lines, broken promises. Moddule's ETA IQ returns a trust-scored prediction you can actually plan against.
Every wrong ETA has a price tag.
Schedule reliability across the global ocean freight network has fallen to less than 60 per cent. That means four out of every ten shipments miss their original prediction, with every conflicting ETA costs someone, somewhere, real money.
Global ocean freight schedule reliability. Four of every ten shipments miss their original ETA.
Conflicting ETAs carried by a single shipment at any given time across aggregators, carriers, and AIS feeds.
Annual demurrage and detention exposure for a single enterprise shipper running on bad predictions.
The amount of inventory carrying cost that is safety stock, which is a direct tax on ETA unreliability.
Four stages from raw signal to trusted action.
ETA IQ sits above your existing visibility and tracking tools, consuming their output, validating it against context, and returning a single trust-scored prediction you can plan against.
Ingest 4-6 sources per shipment
Aggregators, 3PL feeds, carrier direct, port community systems, AIS vessel intel. API push, SFTP, or file-based. You choose what to input.
Enrich with context signals
Carrier performance history, port congestion, route variability, weather, transshipment risk. The signals that actually move predictions.
Score every prediction
A dynamic 0-100 trust score per source. It's calculated for the specific carrier, port, trade lane, and conditions. It's not a static reputation.
Deliver a blended ETA you can plan against
The result is a confidence-weighted prediction via RESTful API at 15, 10 and 5 days out. Actuals feed back into the model. The accuracy compounds.
Not another tracker.
An intelligence layer.
ETA IQ isn't a visibility platform, a data aggregator, or a prediction model competing in a crowded field. It's the layer that sits above all of them and does the scoring, validating, and selecting.
Trust Scoring
A confidence metric (0-100%) that allows ops teams to prioritise action on low-confidence shipments, not chase every alert.
Blended Prediction
A confidence-weighted ETA combining the highest-trust inputs. Not a simple average, but a context-aware selection of the prediction most likely to be right.
Dynamic Switching
Change which provider is trusted at shipment, carrier, port, or trade-lane level. Never locked to a single global source. Always the right provider for the right context.
Provider Recommendation
Dynamic selection of the most accurate provider for a specific shipment at a specific point in the journey. The engine tells you whose signal to take.
Continuous Learning
As actual milestones confirm, the model retrains. Accuracy compounds for specific carriers, routes, and trade lanes over time. The more you ship, the sharper it gets.
Carrier & Route Benchmarking
Data-driven insight into where predictions consistently fail and why, benchmarked across providers. Leverage for procurement, ammunition for ops.
The intelligence layer between seeing it and acting on it.
ETA IQ is Layer 2 of a three-layer architecture. It consumes normalised events from the Visibility Platform and fuels decisioning in Moddule OS. Each layer earns the right to exist because the one below it already does.
Moddule OS
Turns trust-scored predictions into coordinated action across TMS, WMS, ERP and customer comms, automatically, within human-defined guardrails.
ETA IQ
Trust-scores every prediction across providers. Returns a blended, confidence-weighted ETA. Continuously learns from actuals.
Visibility Platform
Normalises shipment events across carriers, modes and systems. The unified data model OS orchestrates against.
Built for the operators who feel the cost of a wrong prediction.
Enterprise logistics teams managing complex, multi-carrier supply chains know the financial impact of poor ETA accuracy isn't abstract, it's on the P&L every quarter.
Provider-agnostic intelligence across your entire network.
Benchmark provider accuracy by trade lane and carrier. Turn that data into procurement leverage and into a value-add you can offer your own customers.
Plan against a prediction you can actually trust.
For fast-fashion, automotive JIT, or seasonal retail, late arrivals compound into markdowns, line stops, and emergency freight. Higher confidence means tighter plans.
Related resources
The Supply Chain Planning Problem Behind Your Inventory
Neil Marchant has spent 25 years in consumer goods planning. His verdict: your inventory problem is really a planning problem.
.png)
Great Storefront, Broken Supply Chain: The eCommerce Growth Trap
Most eComm brands overlook their supply chain, when it's really availability, forecasting, and data that drive growth.
- The four phases from data foundation to visible proof
- How to pick one baseline metric and agree on it on day one
- Three safeguards that protect your investment
The 90-Day Path to Proof for AI Logistics Tech
Run an AI logistics proof of concept that proves value: one baseline metric, four phases, and what you should be holding on day 90.
Frequently asked questions
Practical questions from operations, IT and procurement teams interested in ETA IQ. For anything not covered here, our team will walk you through it on a call.
No. ETA IQ sits above your existing stack. It consumes predictions from whatever aggregators, carrier feeds, and AIS providers you already use, and returns a trust-scored output. The more sources, the sharper it gets.
It's dynamically recalculated for each shipment based on carrier, vessel, port pair, trade lane, conditions, and the provider's historical accuracy in that exact context. Not a static provider rating.
Data in via API push, SFTP, or file transfer. Predictions out via Moddule's RESTful API, documented on Swagger. For pilots, we deliver at three checkpoints, typically 15, 10, and 5 days before ETA.
The commercial bar is measurably better than carrier-provided ETAs. For deeper integration scenarios feeding automated decisions, the target is 95% accuracy at 10 days out. Accuracy compounds as the model ingests more operational data.
Visibility captures and normalizes shipment events. It is the foundational data layer. ETA IQ is the intelligence layer above it. You can run ETA IQ on Visibility data, or on data from other visibility tools you already have
The current focus is ocean, as it is the most universal pain point. But the engine is milestone-agnostic. The same architecture applies to air, rail, and across departure, transshipment, customs, and dwell events.
Stop asking which ETA to trust. Start shipping against the right one.
Let's talk about your shipment volume, your providers, and what a pilot could look like.