When the Default Response Takes Too Long
When a load drops, most dispatch floors begin by calling through the carrier list. Senior dispatchers know which carriers usually have trucks nearby. That knowledge is valuable and grounded in experience. But "usually" carries significant weight. Carrier availability changes continuously rather than remaining fixed. The carrier that reliably ran the Pune-to-Mumbai lane last quarter may have shifted its fleet in response to this week's demand. A call made at 10 AM often relies on information from the last interaction, which may be two weeks old.
Empty miles, also known as deadhead, result directly from the time a disruption remains open before re-assignment. A truck receiving a replacement load within 10 minutes of cancellation remains at the dock. If the assignment arrives 40 minutes later, that truck is already three hours out of position without a paying load.
The Cost Missing From the Invoice
The direct cost of empty miles is clear. A carrier operating 20% deadhead across a 100,000-mile operation travels 20,000 miles without revenue while still paying full fuel and maintenance costs. For mid-market carriers with thin margins, that figure creates a genuine problem that builds throughout the year.
The broker absorbs a less visible cost. When a load remains uncovered for an extended period, two events occur at once. The shipper's pickup window begins closing, while the dispatcher handling recovery loses the ability to oversee the rest of the board normally. Recovery costs include more than phone time. They also include monitoring attention pulled away from 29 other active assignments.
Before building MatchLog, we saw this pattern repeatedly in early discussions with freight operations teams. One unplanned cancellation at 9 AM could create two to three hours of secondary disruption, since the dispatcher handling it was no longer available for regular board oversight.
Where Available Capacity Really Exists
Public load boards display posted capacity, usually from carriers that could not fill it through their regular relationships. They are useful for spot capacity on lanes you do not operate often. For a mid-market broker mainly needing dependable re-assignment on familiar lanes after a regular carrier cancels, however, the board is often the wrong tool. Better visibility into capacity already accessible through existing carrier relationships is the better fit.
In our work with early-access operations teams, the most dependable replacement capacity generally appears in carrier network data and historical lane patterns. A carrier running the Delhi-to-Jaipur lane on Tuesdays and Thursdays will likely have a truck near Delhi on Monday evening. That pattern does not show up in a DAT posting. It exists in TMS history and in dispatchers' knowledge.
The difficulty is retrieving that knowledge under pressure. The dispatcher must recall historical data, verify current carrier status, and make phone contact at once. These three steps are sequential, and each consumes time and attention.
What Real-Time Matching Requires
Real-time matching is mainly a data freshness challenge, not a matching algorithm challenge. Even the strongest matching logic produces poor results when the carrier position data behind it is four hours old.
Most TMS platforms refresh carrier positions periodically rather than continuously. That suits planning, but not disruption response. The important question then is not "where was this carrier this morning" but "where is this carrier right now and can they make this window."
MatchLog takes in carrier telemetry through ELD integrations and carrier API feeds, with the clear aim of keeping position data current enough for same-day re-assignment decisions. Much of our engineering time has focused on stale data, ELD connectivity gaps, and position reports that conflict with carrier-reported status. A sound data layer must come before sound matching. The algorithm is the smaller challenge.
MatchLog: Scope and Limits
MatchLog is neither a load board nor a carrier discovery tool. It operates within your existing carrier network, bringing forward better-fit candidates faster from carriers with whom you already work. Recommendations use current position and historical lane performance. If your network has limited coverage on a lane, MatchLog will show that gap. It will not fill the gap with unknown carriers.
We are also clear about what "faster matching" means in practice. Across our 3-customer early-access cohort over a 60-day window, we observed that loads remained uncovered for less time after an initial disruption. Yet shorter coverage time and lower fleet-level deadhead are connected without being the same problem. Moving from one outcome to the other depends on variables far beyond a dispatch tool's control.
The 10-Minute Window
The number that repeatedly appears in early-access feedback is 10 minutes. Disruptions detected and covered within roughly 10 minutes usually do not cascade into secondary disruptions. When they remain open longer, they begin consuming more dispatcher attention and creating downstream effects across the rest of the board.
Real-time matching is not about finding the perfect carrier assignment. It is about putting a good assignment before a dispatcher quickly enough for a decision before the window closes. Surfacing candidates quickly matters more than algorithmic optimality, since a 98% optimal match confirmed in 5 minutes is better than a 100% optimal match confirmed in 45.
That perspective guides MatchLog. The dispatcher remains the decision-maker. Our role is to shorten the time between a disruption and a confirmable option appearing on their screen. The rest follows.