I've been running fleet operations for a mid-size logistics company for about seven years now. 52 trucks, 14 vans, mostly regional hauls across the Midwest and Southeast. Nothing fancy.
A couple years ago our CFO dropped a spreadsheet on my desk showing total fleet costs per mile had gone up 23% in two years. Fuel was part of it, sure. But fuel wasn't the whole story. So I spent the next 12 months tracking where every dollar actually went. What I found surprised me, and I figure it might help someone here avoid the same mistakes.
I'll share the numbers and what we changed. Some of this is basic, some isn't.
Here's what was causing most of them:
This was the single biggest change. We used to service every truck at fixed intervals — oil every 15,000 miles, brakes inspected every 30,000, coolant flush once a year. Simple, easy to schedule, and completely wrong for half our fleet.
Trucks running short-haul urban routes with constant stop-and-go put way more stress on brakes and transmissions than highway haulers covering the same distance. One of our city delivery trucks needed brake work at 18,000 miles. A highway truck on the same schedule? Brakes were still at 60% life at 30,000.
Once we started scheduling based on actual component data — oil analysis results, brake wear measurements, coolant condition, filter pressure differentials — our maintenance costs dropped about 19% in the first year. We were servicing some trucks more often and some less often, but every service was actually needed instead of just calendar-driven.
Started using telematics data for early fault detection
This is where things got interesting. We were already collecting telematics data through our ELD provider, but we weren't doing anything smart with it. It was basically just GPS tracking and hours-of-service compliance.
We added a diagnostic layer on top — specifically Intangles' predictive maintenance platform — that pulls live data from the engine ECU, transmission, brakes, battery, and other systems. The software flags anomalies before they become failures. A coolant temp that's trending 8 degrees higher than the same truck's 90-day average? That gets flagged. A battery that's losing voltage capacity faster than expected? Flagged.
In the first six months, we caught 23 issues that would have been roadside breakdowns. At $2,800 average per event, that's roughly $64,000 in avoided costs. The platform paid for itself in about two months.
I'll be honest — Samsara, Geotab, and a few others offer similar monitoring. We went with Intangles because their fault prediction was more granular for our mixed-brand fleet (we run Freightliner, International, and some older Kenworths) and it worked with our existing hardware without needing a full rip-and-replace. Your situation might be different.
Fuel management — beyond just watching the pump price
Everyone tracks fuel spend. Not enough people track fuel efficiency at the individual vehicle and route level. When we started doing per-truck, per-route fuel analysis, we found:
Tire management is boring but the ROI is real
Nobody wants to talk about tires. I get it. But we were spending about $68,000/year on tires across the fleet, and a lot of that was avoidable.
What changed:
Happy to answer questions. I know every fleet is different but the cost patterns are surprisingly similar once you look at the data.
A couple years ago our CFO dropped a spreadsheet on my desk showing total fleet costs per mile had gone up 23% in two years. Fuel was part of it, sure. But fuel wasn't the whole story. So I spent the next 12 months tracking where every dollar actually went. What I found surprised me, and I figure it might help someone here avoid the same mistakes.
I'll share the numbers and what we changed. Some of this is basic, some isn't.
Where the money was actually going
When I broke everything down per vehicle per month, here's roughly how our costs split:- Fuel: 38%
- Maintenance and repairs: 22%
- Insurance: 14%
- Depreciation: 12%
- Tires: 6%
- Driver downtime and overtime (caused by breakdowns): 5%
- Miscellaneous (tolls, permits, admin): 3%
The breakdown problem (literally)
In 2023, we had 74 unplanned breakdowns across the fleet. Average cost per event — including towing, emergency repair, lost delivery time, and overtime for covering the route — came out to about $2,800. That's roughly $207,000 in a year just from stuff breaking when we didn't expect it.Here's what was causing most of them:
- Cooling system failures: 19 events
- Brake issues caught too late: 14 events
- Electrical and sensor faults: 12 events
- Transmission overheating: 9 events
- Tire blowouts: 8 events
- Other (alternator, AC compressor, fuel system): 12 events
What we changed — and what actually worked
Switched from calendar-based to condition-based maintenanceThis was the single biggest change. We used to service every truck at fixed intervals — oil every 15,000 miles, brakes inspected every 30,000, coolant flush once a year. Simple, easy to schedule, and completely wrong for half our fleet.
Trucks running short-haul urban routes with constant stop-and-go put way more stress on brakes and transmissions than highway haulers covering the same distance. One of our city delivery trucks needed brake work at 18,000 miles. A highway truck on the same schedule? Brakes were still at 60% life at 30,000.
Once we started scheduling based on actual component data — oil analysis results, brake wear measurements, coolant condition, filter pressure differentials — our maintenance costs dropped about 19% in the first year. We were servicing some trucks more often and some less often, but every service was actually needed instead of just calendar-driven.
Started using telematics data for early fault detection
This is where things got interesting. We were already collecting telematics data through our ELD provider, but we weren't doing anything smart with it. It was basically just GPS tracking and hours-of-service compliance.
We added a diagnostic layer on top — specifically Intangles' predictive maintenance platform — that pulls live data from the engine ECU, transmission, brakes, battery, and other systems. The software flags anomalies before they become failures. A coolant temp that's trending 8 degrees higher than the same truck's 90-day average? That gets flagged. A battery that's losing voltage capacity faster than expected? Flagged.
In the first six months, we caught 23 issues that would have been roadside breakdowns. At $2,800 average per event, that's roughly $64,000 in avoided costs. The platform paid for itself in about two months.
I'll be honest — Samsara, Geotab, and a few others offer similar monitoring. We went with Intangles because their fault prediction was more granular for our mixed-brand fleet (we run Freightliner, International, and some older Kenworths) and it worked with our existing hardware without needing a full rip-and-replace. Your situation might be different.
Fuel management — beyond just watching the pump price
Everyone tracks fuel spend. Not enough people track fuel efficiency at the individual vehicle and route level. When we started doing per-truck, per-route fuel analysis, we found:
- 4 trucks consistently burning 15-25% more fuel than identical units on the same routes. Two had injector problems, one had a turbo issue, and one had a driver with a heavy foot and a 40-minute idling habit at every stop.
- Route optimization alone (avoiding specific congested corridors during peak hours) saved about 6% on fuel across our regional fleet.
- Idle time was costing us roughly $1,100/month across the fleet. We set up automated idle alerts and fuel monitoring and that number dropped to about $400/month within 60 days. Most of the improvement came just from drivers knowing it was being tracked.
Tire management is boring but the ROI is real
Nobody wants to talk about tires. I get it. But we were spending about $68,000/year on tires across the fleet, and a lot of that was avoidable.
What changed:
- Added tire pressure checks to the digital pre-trip inspection (mandatory, with photo confirmation)
- Switched from calendar-based rotation to wear-pattern-based rotation
- Started tracking tire life by position, brand, and route type
The numbers after 12 months
After implementing everything above, here's where we landed year-over-year:- Unplanned breakdowns: 74 → 41 (45% reduction)
- Average cost per breakdown event: $2,800 → $2,100 (faster detection = smaller repairs)
- Total maintenance spend: down 19%
- Fuel costs: down 11% (even with fuel price increases during that period)
- Tire costs: down 25%
- Vehicle uptime: 89% → 96%
What I'd tell someone starting from scratch
If you're running a fleet and haven't gotten into data-driven maintenance yet, don't try to do everything at once. Start here:- Track your real breakdown costs — not just the repair invoice, but towing, downtime, overtime, missed deliveries. The full picture is always worse than you think.
- Get component-level data flowing — ELD-level GPS tracking isn't enough. You need engine, brake, battery, and transmission data to catch problems early.
- Move to condition-based scheduling — fixed intervals waste money on trucks that don't need service and miss trucks that do.
- Monitor fuel at the truck level — fleet-wide averages hide the outliers that are costing you the most.
Happy to answer questions. I know every fleet is different but the cost patterns are surprisingly similar once you look at the data.