I spent four years as a logistics director at a Midwest food distributor before starting Routevein, and every year we ran the private fleet versus outsourcing analysis. Every year we came to roughly the same conclusion through roughly the same spreadsheet. We compared driver wages and benefits against carrier rates. We modeled truck acquisition cost versus per-mile brokered rates. We factored in liability insurance, DOT compliance overhead, and dispatch headcount. And every year, the math was close enough that the decision defaulted to "stay the course."
What that analysis never included — what most make-vs-buy analyses for private fleets still don't include — is the routing efficiency gap between what a mid-market private fleet achieves manually and what an optimized fleet achieves. That gap is large enough to change the outcome of the comparison. Here's how to actually account for it.
What Standard Fleet Cost Models Miss
A typical private fleet cost model looks at total cost of ownership per mile: driver wages, fuel, depreciation, insurance, maintenance, and dispatch overhead. Then it compares that against the fully-loaded cost of outsourcing the same volume to a 3PL or spot market. The breakeven question is usually around $2.80–$3.20 per loaded mile for private fleets versus $3.50–$4.50 per mile for outsourced coverage on similar lanes, depending on geography and volume density.
What these models assume is that the private fleet's actual miles driven equal the theoretical miles required to serve the stop manifest. They don't. Manual routing consistently produces routes with 12–18% more miles than optimally routed alternatives, according to operational data we've gathered from fleet conversations. That inefficiency has two components:
- Deadhead miles: Empty miles driven between delivery legs, or returning to the yard without a reload. A fleet running 18% deadhead on a 150-mile average daily route is burning roughly 27 miles per truck per day in fuel and driver time with zero revenue attached.
- Stop sequencing inefficiency: Even when drivers are loaded, suboptimal stop sequences mean more total distance traveled. A driver covering 22 stops manually sequenced often drives the same geography 8–11% less efficiently than an AI-optimized sequence of the same stops.
Together, these two factors mean your actual cost per effective mile — the cost per mile that moves product, not just moves trucks — is significantly higher than your model shows. The private fleet looks more expensive than it really needs to be, and the outsourcing comparison looks artificially favorable.
Running the Routing Efficiency Adjustment
To correct for this, add a routing efficiency adjustment to your private fleet cost model. The mechanics are straightforward.
Take your fleet's current total annual miles. Subtract an estimate of your deadhead percentage (look at your ELD data — most carriers can pull this from their TMS or Samsara dashboard). Then apply a sequencing efficiency factor. If you're running manual dispatch without optimization software, assume 10–15% sequencing waste is reasonable. If you have basic GPS routing but no AI re-sequencing, assume 7–10%.
The combined routing waste for a typical 50-truck fleet running 250 days per year at 120 miles per truck per day looks like this:
| Factor | Estimate | Miles/Year |
|---|---|---|
| Total fleet miles | 50 trucks × 120 mi/day × 250 days | 1,500,000 |
| Deadhead (15%) | 225,000 miles | 225,000 |
| Sequencing waste (12%) | 180,000 miles | 180,000 |
| Total recoverable miles | Combined inefficiency | 405,000 |
At $0.55 per mile in direct fuel and variable costs, 405,000 recoverable miles represents $222,750 in annual waste — before factoring in driver hours, which are at least as significant. That's not a rounding error in a fleet P&L. It's a material cost pool that the standard make-vs-buy model treats as fixed and unavoidable.
The Outsourcing Counterargument and Its Limits
When logistics managers present the outsourcing case, the implied argument is that a 3PL has already solved the routing efficiency problem at scale — that their larger network means better lane density, lower deadhead, and more efficient sequencing on your freight. For some lanes, in some geographies, that's true. A large 3PL servicing a dense urban corridor can offer genuine density advantages that a 50-truck private fleet can't replicate.
But mid-market shippers rarely have freight that fits neatly into a 3PL's existing network. Regional food distributors, building materials suppliers, and specialty industrial distributors often have service requirements — specific delivery windows, temperature control, white-glove stop protocols, customer-facing driver accountability — that don't port cleanly to a brokered carrier relationship. When those service requirements are written into carrier contracts, the quoted rate goes up significantly. The outsourcing math that looked attractive at $3.80/mile often reaches $4.20–$4.60/mile once service level requirements are fully specified.
We've also seen carriers deprioritize mid-market volume during tight capacity periods. A shipper with 50 trucks of consistent volume has more operational reliability than a shipper dependent on spot market availability during peak season. That reliability premium rarely shows up in the standard analysis.
What Route Optimization Does to the Private Fleet Decision
Here's the adjustment that changes the analysis: if a private fleet deploys AI route optimization and closes even 60% of the routing efficiency gap, the cost picture shifts materially.
Using the same 50-truck example: closing 60% of the 405,000 recoverable miles means eliminating approximately 243,000 inefficient miles per year. At $0.55 variable cost per mile plus proportional driver time, that's roughly $133,650 in direct savings — before accounting for OTD improvement, which typically carries its own penalty reduction value of $85–$140 per avoided missed delivery window.
A fleet running Routevein's Growth plan at $799/month spends $9,588/year on software. The net annual value of routing optimization — conservative estimate — exceeds $120,000 on a 50-truck fleet. That changes the breakeven point in the make-vs-buy analysis by roughly $0.20–$0.30 per loaded mile, which is often enough to flip the conclusion.
"The standard fleet analysis compares a well-run 3PL against a poorly-optimized private fleet. That's not an apples-to-apples comparison. The right comparison is a well-run 3PL against an optimized private fleet."
Timing and Organizational Readiness
The routing efficiency adjustment doesn't always favor keeping the private fleet. There are legitimate scenarios where outsourcing is the right answer: fleets with highly volatile volume profiles, shippers entering new geographies without existing lane data, operations where dedicated lane density is too thin to optimize. The point isn't to prove private fleet always wins. It's to ensure the analysis is complete before the decision is made.
In my experience, the decision to outsource made on an incomplete analysis tends to surface its costs within 12–18 months. Service quality mismatches, capacity gaps during seasonal peaks, and the loss of direct customer relationships all show up as follow-on problems that weren't priced into the original comparison. By then, rebuilding a private fleet operation is expensive and slow.
Equally, the decision to retain a private fleet without investing in routing optimization just defers the efficiency problem. You've paid the capital and labor cost of owning the fleet but haven't captured the operational savings that would justify that investment.
Practical Next Steps
If you're running this analysis now, here's what I'd add to your model:
- Pull your ELD data and calculate your actual deadhead percentage for the last 90 days. Most Samsara dashboards or McLeod TMS reports can surface this directly.
- Estimate your stop sequencing efficiency. Run your last week of routes through a free routing tool and compare total distance. The gap is usually visible in under an hour.
- Apply a conservative 50–60% recovery assumption to the combined inefficiency figure — don't assume you'll capture all of it, because no optimization captures everything.
- Add the recovered cost to your private fleet column before comparing against outsourcing rates.
The make-vs-buy decision for a private fleet is genuinely complex. But it should be complex in all the right ways — including the routing efficiency variable that most models quietly leave out of the math.
