White paper
Driver Retention is a Data Problem
by Rapid Relay Research Team·June 19, 2026·20 min read·Data Mining
TL;DR
Driver turnover at large long-haul carriers averages 94% a year, costing the industry an estimated $18.7 billion. The paper shows these exits are operational rather than a pay problem, and predictable from data carriers already collect.
Key findings
- Large long-haul carriers lose roughly 94% of their drivers every year, a churn rate the industry has quietly accepted as a cost of doing business rather than a crisis to solve.
- Replacing a single driver costs about $12,799, and the industry spends an estimated $18.7 billion a year on departures that were largely preventable.
- Route length, home time frequency, and asset utilization explain more of the variation in turnover than pay raises do, which makes turnover an operational problem rather than a compensation problem.
- Most turnover is drivers switching carriers, not leaving trucking, so the real failure is competitive: carriers are losing drivers to operators who run better lanes.
- Relay networks restructure the three variables that most reliably predict driver exits, the structural fix the paper builds toward, which is why short-haul, regional, and dedicated operations retain drivers far better than long-haul OTR.
By the numbers
94%
Long-haul driver turnover at large carriers
Source: ATRI, Evolving Truck Driver Demographics (Aug. 2025); ATA historical avg 1996–2023
48%
Industry-average driver turnover, 2024
Source: ATRI, Operational Costs of Trucking (July 2025)
$12,799
Estimated cost of replacing a single driver
Source: PDA / Conversion Interactive Agency Snapshot, Feb. 2025
$18.7B
Estimated annual industry cost of driver replacement
Source: Rapid Relay Technologies calculation (BLS + ATRI data)
~80%
Of sub-500-mile-LOH fleets report turnover below 50%
Source: TMW Systems / FleetOwner carrier survey
49.1%
Of job-seeking drivers cite the need for consistent miles
Source: PDA, Fall 2024 Driver Survey
In their own words
They leave when the job doesn't match what they were told, when no one checks in, when the money isn't there, and when they feel like just another seat to fill.
If you think there's a shortage, explain to me why there's 400,000-plus new CDLs issued in this country every single year. We have a driver turnover problem, with over 90% turnover in the long-haul trucking industry.
Retention is about more than just pay rates. It is about ensuring drivers can actually earn the money they were promised.
There's a very good chance in 2025 and 2026 that we'll get the economy rolling again. Freight will start to pick up, at which point the big carriers, as they have in the past, struggle to recruit and retain drivers.
Methodology
Every figure is drawn from public sources, including ATRI, the ATA, PDA / Conversion Interactive Agency, UGPTI, TMW Systems via FleetOwner, the National Academies, and the BLS, and each is cited in full in the paper. The analysis can be checked independently.
The headline $18.7 billion estimate scales the problem against broad BLS employment. Narrowed to the roughly 2.2 million heavy and tractor-trailer drivers the BLS tracks, the annual cost is closer to $13.6 billion, where a 10-point improvement in retention would save about $1.4 billion. Both are order-of-magnitude estimates rather than forecasts.
FAQs
Isn't driver turnover a pay problem? Why isn't paying drivers more the answer?
Industry survey data says otherwise. In the PDA Fall 2024 Driver Survey, 81.9% of job-seeking drivers cited predictable pay as a reason for switching, but critically, the 2024 PDA Snapshot found that 60% of drivers who complained about compensation specifically cited lack of miles (not the per-mile rate) as the root cause. Drivers aren't leaving for higher per-mile pay; they're leaving because their carrier's operations aren't generating consistent mileage. The problem is operational performance, not compensation design, and the paper walks through the survey data behind that conclusion.
Is the 94% turnover rate a typo? That can't be right.
It's not a typo. The American Trucking Associations has tracked turnover at large truckload carriers (those earning $30M+ in annual revenue) at an average of 92.7% from 1996 to 2023. ATRI's August 2025 report places long-haul OTR turnover at 94% at large carriers. For every ten drivers a major long-haul carrier employs at the start of the year, nine are typically gone before it ends.
If retention is predictable, why isn't every carrier doing this?
The data exists in every mid-to-large carrier's TMS and ELD systems: miles driven vs. expected, days from home, load acceptance rates, communication patterns. What's missing is the synthesis layer that connects those signals into a continuously-updated retention risk score at the driver and lane level. The industry has been measuring retention with annual fleet-wide rates (a lagging indicator) rather than continuous lane-level analytics. Carriers that move first will have a structural information advantage.
Doesn't fixing this require expensive infrastructure (relay points, terminals, depots)?
Not necessarily. Asset-light relay orchestration (coordinating handoffs across existing terminals, drop yards, customer sites, and neutral relay points) achieves the same structural retention outcomes (shorter hauls, more frequent home time, higher utilization) without new capital investment. The operational redesign should precede any asset decision. Carriers that first identify which lanes have retention-degrading profiles, then model the relay configuration, often find the required handoff infrastructure already exists in their network. The report works through this objection in detail.
How does this paper define the difference between an industry shortage and a retention problem?
The ATA itself has been explicit on this point: the turnover figure measures churn within the industry, not attrition from it. The majority of drivers who register as 'turnover' are not leaving trucking; they're leaving one carrier for another. An industry framing this as a shortage will try to recruit its way out. An industry framing it as a competitive retention failure will look for the operational conditions that drive exits and address them. The two framings lead to very different interventions.