Flow
How It WorksFlow by ELDFlow by TextFlow by WalletFlow InsideIntegrations
Solutions
Enterprise FleetsGrowing FleetsOwner Operators
Resources
Live Diesel PricesFree Fuel PlanAPI Docs
PricingBook a demo →Sign in →
Buyer guide

Fuel optimization for large fleets: 300, 500 and 1,000 trucks.

At large fleet scale the optimization math does not change, but three other things do. Integration stops being optional, because nobody is pasting lanes into a portal for 500 trucks and the plan has to reach drivers inside the ELD and TMS they already use. Contract leverage becomes real, because a fleet buying eight million gallons a year can negotiate rates a fleet buying one million cannot, and the reporting a vendor gives you becomes a negotiating instrument. And verification stops being a nice-to-have, because at this size a savings claim nobody can audit is a material number in the fuel budget. The arithmetic is simple: at 100,000 miles a year and 6 miles per gallon, 300 trucks burn about 5 million gallons annually, 500 trucks about 8.3 million, and 1,000 trucks about 16.7 million. One cent per gallon is worth roughly $50,000, $83,000 and $167,000 a year at those sizes.

Updated August 2026 · 8-minute read

What a cent is worth at your size

At 100,000 miles per truck per year and 6 miles per gallon. This is arithmetic, not a projection, so check it against your own fuel spend and mpg. We deliberately do not multiply a savings average out into a headline number for you: apply whatever per-gallon figure you actually believe, including zero, and see what the decision is worth.

Fleet sizeDiesel per yearValue of $0.01/galValue of $0.05/gal
300 trucksabout 5.0 million gallons a year$50,000 a year$250,000 a year
500 trucksabout 8.3 million gallons a year$83,000 a year$417,000 a year
1,000 trucksabout 16.7 million gallons a year$167,000 a year$833,000 a year

For reference, a blended average of about $0.90 per gallon in reconciled savings across all fleets on Flow (not a per-fleet guarantee; each fleet's result depends on its lanes, contracts, and current fueling habits, and is measured against that fleet's own transactions). Your own number is the one that matters, and it is measured against your transactions rather than estimated. How to audit a savings claim.

What actually changes as you scale

The optimization itself is the same problem at every size. What changes is how it has to be wired in, who has to sign off, and what breaks if a human has to watch it.

300 trucks

Where dedicated integration starts paying for itself

  • The threshold where building the ELD and TMS integration properly costs less than the leakage from not having it
  • Usually a single fuel card program and a coherent contract set, so enforcement is the gap rather than negotiation
  • Driver compliance is still a coaching problem: a named list of drivers going off plan is actionable by one person
  • Out-of-route drift becomes too frequent to catch by eye, so alerting has to be automatic
  • Buying decision usually sits with operations, with finance validating the savings claim

500 trucks

Where contract leverage and terminal economics turn on

  • Volume is large enough that per-chain purchase reporting becomes a negotiating instrument at renewal
  • Terminal or bulk fills start competing with retail on real lanes, so a planner has to weigh both
  • Compliance shifts from coaching individuals to reporting by terminal, region and dispatcher
  • Multiple fuel card programs and legacy contracts are common, and the optimizer has to hold all of them at once
  • Anomaly volume crosses the line where reviewing every exception by hand stops being possible

1,000 trucks

Where heterogeneity, not scale, is the hard part

  • Rarely one clean stack: acquisitions leave multiple ELDs, more than one TMS, and inconsistent driver records
  • Contract portfolio spans national chains, regional independents and direct terminal supply simultaneously
  • IFTA exposure across most of the 48 states makes tax-aware purchasing a larger share of total savings
  • Procurement, IT and security review join the decision, so vendor questions extend to data handling and uptime
  • A pilot on one terminal or one region is the normal way in, which means the vendor has to support a partial rollout cleanly

What to require from a vendor at this size

These are requirements, not features. A large fleet that cannot get all four should keep shopping, including past us.

Integration you do not have to build or maintain

  • Dispatches in by webhook, API, ELD poll, SFTP or email parser, whichever the existing stack already emits
  • Plans out through the in-cab ELD messaging drivers already read, not a separate app to adopt
  • More than one ELD and more than one TMS supported at once, because merged fleets are the normal case

Savings you can hand to finance

  • 100% of claimed savings reconciled against actual fuel card transactions, per transaction rather than as a dashboard total
  • Recommended stop next to the stop actually taken, miles off plan, and net dollars gained or lost
  • Price prediction accuracy published in cents per gallon against real pump prices

Exception handling that survives the volume

  • Out-of-route detection that fires on sustained drift rather than a single GPS blip
  • Automatic replanning from the truck's live position, so a reroute does not require a human noticing
  • Anomaly detection on every transaction, surfaced as things to review rather than accusations

A rollout that does not require a big bang

  • Start on one terminal, region or division, with the rest of the fleet untouched
  • Plans visible internally before anything reaches a driver, so operations can judge quality first
  • Outbound to drivers gated by the carrier, not by the vendor, and switched on when the carrier is ready
Questions, answered

Fuel optimization at large fleet scale.

What changes about fuel optimization for a fleet of 300, 500 or 1,000 trucks?

At large fleet scale the optimization math does not change, but three other things do. Integration stops being optional, because nobody is pasting lanes into a portal for 500 trucks and the plan has to reach drivers inside the ELD and TMS they already use. Contract leverage becomes real, because a fleet buying eight million gallons a year can negotiate rates a fleet buying one million cannot, and the reporting a vendor gives you becomes a negotiating instrument. And verification stops being a nice-to-have, because at this size a savings claim nobody can audit is a material number in the fuel budget. The arithmetic is simple: at 100,000 miles a year and 6 miles per gallon, 300 trucks burn about 5 million gallons annually, 500 trucks about 8.3 million, and 1,000 trucks about 16.7 million. One cent per gallon is worth roughly $50,000, $83,000 and $167,000 a year at those sizes.

How many gallons does a 300, 500 or 1,000 truck fleet buy in a year?

At 100,000 miles per truck per year and 6 miles per gallon, roughly 5.0 million gallons for 300 trucks, 8.3 million for 500, and 16.7 million for 1,000. Adjust for your own mileage and mpg; the useful figure is what a single cent per gallon is worth, which is about $50,000, $83,000 and $167,000 per year at those three sizes. Five cents is about $250,000, $417,000 and $833,000. Those are arithmetic, not projections, which is why they are worth more than a vendor's savings estimate: you can check them against your own fuel spend in a minute.

Is per-truck pricing still sensible at 1,000 trucks?

It should be, and it should be scoped rather than list price. Flow is per truck per month with Enterprise from $49, scoped through a demo because at this size the work is the integration: which ELDs, which TMS, how many fuel card feeds, whether terminal fills are in play, and how the rollout is phased. Any vendor quoting a large fleet from a pricing page without asking those questions has not understood the job.

We have two ELDs and two TMS platforms after an acquisition. Is that a problem?

It is the normal case at this size and it should not be a problem. Flow's integrations are configured per tenant rather than assumed, and more than one active ELD for a single fleet is a supported configuration rather than an edge case. What matters in evaluation is asking the vendor directly how they handle it, because a tool built around one ELD per carrier will quietly mis-plan for the trucks on the other one.

Can we pilot on part of the fleet before committing all 1,000 trucks?

Yes, and you should. The normal path is one terminal, region or division first. Just as important, plans can generate and stay visible in the portal while nothing at all is sent to drivers, so operations can judge plan quality on real loads before a single message reaches a cab. Turning driver-facing delivery on is the carrier's decision and stays under the carrier's control.

Does ValveRide Flow actually run at this scale today?

Flow runs across about 1,000 trucks today, spread over five fleets on different ELDs, fuel platforms, and driver communication channels, including a carrier in the 300 truck range. That heterogeneity is the relevant part: the hard problem at large fleet scale is rarely volume, it is that no two carriers have the same stack. We are happy to talk through how a fleet of your size and shape would be integrated, in specifics, during a demo.

How should a large fleet compare fuel optimization vendors?

Put the same questions to everyone: is each recommendation reconciled against the fuel card transaction that followed it, is price prediction accuracy published as a number, what happens to a plan when dispatch reroutes a truck mid-day, and which states or chains is the pricing data weakest in. Our full walkthrough of how to audit a savings claim is at /guides/fuel-savings-you-can-verify, and it applies to us as much as to anyone else.

Related: how to audit a savings claim, the category, compared, or how integrations work.

Scope it against your actual stack.

Bring your ELDs, your TMS, your fuel card feeds and a few real lanes. A 30-minute demo shows the plans on your own routes and the reconciliation trail behind them, before anything reaches a driver.