Our data shows that OpenTrack ETAs are more accurate than those provided by one of the other data sources alone.
Our key principles:
1. We standardize on berthing. Steamship Line ETAs are sometimes to anchorage, sometimes to berthing. This leads to confusion and operational mistakes. OpenTrack’s ETA is always to vessel berthing, giving you the clarity you need to operate a solid process and make confident decisions.
2. We gather ETAs from more data sources. Sometimes the Terminal provides the most accurate ETA, and sometimes it’s the rail carrier. Sometimes it comes from a prediction based on the vessel’s GPS location.
3. We choose the best ETA. Over the years, we’ve worked with hundreds of operators and crunched a ton of data to create intelligent ETA selection strategies designed to produce the most accurate result and mimic the decisions a seasoned operator would make. And we’re constantly improving.
Your operations depend on accurately knowing when shipments will arrive. Inaccurate ETAs are costing your organization time and money. OpenTrack’s ETAs can help.
Comparing ETA accuracy by source
Let’s say you had a shipment going from Yantian to Long Beach. The vessel actually berthed on April 10 at 6:03am. That’s your ATA. Now let’s look at historical performance:
10 days before ATA, the Steamship Line was reporting an ETA of April 5 at 4:30pm, 5 days off from ATA.
While OpenTrack was reporting an ETA of April 10 at 5:30am.
5 days before ATA, the Steamship Line was reporting an ETA of April 6 at 8:30am, 4 days off fromATA.
While OpenTrack was reporting an ETA of April 10 at 5:30am.
2 days before ATA, the Steamship Line was reporting an ETA of April 9 at 8:30pm, 1 day off from ATA.
While OpenTrack was reporting an ETA of April 10 at 5:30am.
If you prefer tables, that looks something like this:
Days before ATA | SSL’s ETA | SSL’s accuracy | OpenTrack’s ETA | OpenTrack’s accuracy |
10 | April 5 at 4:30pm | 5 days off | April 10 at 5:30am | 33 mins off |
5 | April 6 at 8:30am | 4 days off | April 10 at 5:30am | 33 mins off |
2 | April 9 at 8:30pm | 1 day off | April 10 at 5:30am | 33 mins off |
It’s not uncommon to see Steamship Line ETAs that are inaccurate by multiple days like this. Wouldn’t you rather have the most accurate ETA from day one?
And that’s just one example. On average, across all shipments we track, OpenTrack’s ETAs are more accurate than the Steamship Line’s every step of the way:
Why ETA accuracy matters
We know you know, but it never hurts to restate the importance of accurate ETAs.
Inaccurate ETAs from Steamship Lines alone | Accurate ETAs from OpenTrack |
🚫 More last-minute scrambling when the arrival date is different than you expected | ✅ More advance planning, less stress, and fewer costly mistakes |
🚫 Always be second-guessing whether your ETA is to anchorage or berthing | ✅ Always know that your ETA is to vessel berthing |
🚫 Only use one source for your ETAs | ✅ Get ETAs from multiple sources |
🚫 Force your team to manage data directly from the source | ✅ Get a clean, standardized, reliable data feed your team can use right away |
Our process to evaluate and improve ETA accuracy
We’re intense about measuring and improving our ETA accuracy. The result is that OpenTrack’s ETAs are better overall than any one source alone.
Here’s a window into how that process looks for us:
Every day thousands of vessels berth at their terminals of discharge, registering an actual time of arrival (ATA) in OpenTrack.
When that happens, we look back at what ETA we were reporting not only on the day of arrival, but on each day going backward in time.
Our goal is: On every single day leading up to arrival, we want to report an ETA that is as close as possible to the ultimate ATA.
Every time an ETA is predicted, whether by a data provider (Steamship Line, Terminal, AIS Provider), or by OpenTrack, we make a note of the prediction. That way we can compare the performance of each data provider on each day leading up to the ATA.
In order to improve, we regularly examine and eliminate cases where we’re most inaccurate. We also measure how we’re doing in total.
How we choose the best ETA
We report the Mother Vessel ETA (at Terminal) for all ocean containers, at every stage of the container voyage.
First, if two or more sources agree on one date, then we take that one.
But if sources don’t agree, then we choose the ETA based on the credibility of the source. We rank the credibility of our sources based on each source’s historical ETA accuracy on the container’s current route or lane.
Here’s a breakdown of the data sources and decision strategies used in each phase of the container’s voyage:
Stages | At Origin | On the water | Within 10 days of ETA |
Sources | Steamship Line 🚢 | Steamship Line 🚢 | Steamship Line 🚢 |
How do we decide which ETA to use? | Steamship Line 🚢 | Consensus: If two or more sources agree on one ETA, then we take that one. | Consensus: If two or more sources agree on one ETA, then we take that one. |
Why don’t all containers on this MBL or vessel have the same ETA?
Sometimes containers on the same MBL or vessel may have different ETAs, but this is typically resolved within a few hours. If the differences persist, there may be a good reason for it.
This would be the same if someone were doing manual track and trace: They would update the data container by container, and at any given time containers on same MBL or same vessel could have different ETAs if the ETA is changed partway through.
We don’t automatically apply the same ETA to every container on the same vessel because:
The same vessel may have multiple berthings at different terminals in the same port
The same vessel may have multiple berthings at the same terminal (yes, we’ve seen this happen!)
Sometimes the container’s terminal of discharge may change, so applying its ETA to the whole vessel would give the wrong ETA to every other container on the vessel
We’ve found also that Steamship Lines and Terminals can report bad ETAs temporarily. So if we were to apply the bad ETA to all containers on a vessel, that would be wrong.
We used to do this and learned the hard way that it doesn’t work.
Future developments
We’re constantly reviewing our performance and improving. So the more data we see, the better it gets.
OpenTrack intelligence is only getting better thanks to new machine learning techniques that help us predict ETAs with higher accuracy based on more factors.
We’re always investigating adding new data sources that can contribute to providing more accurate ETAs.