Quick Answer
Four levers, in order of size. Revenue mix moves margin most: shifting capacity from broker-only to a broker-plus-private mix takes net margin from roughly 8-15% toward 20-40% on the shifted portion. Cost per trip is next, and dispatch is the largest controllable line. No-show rate is third and the fastest to fix. Dispatch capacity is fourth and gates the other three, because a fleet that cannot accept trips cannot grow into any of them.
The US NEMT market is around $12.5 billion and growing at roughly 5.5% a year, so the sector is not the constraint. What separates fleets earning well from fleets working hard is a short list, and it is the same list every time.
Lever 1: revenue mix
This is the biggest one and the slowest. Broker trips return 8-15% net for a well-run operator; private pay returns 25-40% and facility contracts 20-35% with better predictability.
The realistic target is not "replace brokers". It is capping broker work at about half of fleet capacity and filling the remainder with private and facility volume, which is the mix most well-run operators settle into.
The work is covered in the six private-ride channels.
Lever 2: cost per trip, and where it actually sits
Fuel and wages are the obvious lines and the hardest to move. Dispatch is the line most operators have never costed properly.
For a fleet running 50 trips a day, in-house dispatch runs roughly $15,000-25,000 a month all in — salary, cover, turnover, software, the manager who backfills. Outsourced runs $4,000-9,000, which is where the frequently quoted 30-50% reduction in total dispatch cost comes from.
The number that matters more is what it does to fixed cost. Outsourced dispatch scales with trip volume; a salaried dispatcher does not shrink in a slow month.
Want NEMT-trained agents answering in your company name?Free operations audit — a written plan within 1 business day.
Get My Free AuditLever 3: no-shows
A no-show is a fully loaded trip cost with no revenue against it, and most fleets carry a rate they have never measured.
Confirming the next day's board is unglamorous and works. The second-order effect is usually larger than the first: the dispatcher stops beginning each morning by rebuilding the run.
Lever 4: dispatch capacity gates everything
Acceptance rate drives broker allocation. A fleet declining trips because dispatch is stretched gets offered fewer, and the decline is invisible until volume has already fallen.
The same constraint caps private growth: a facility that cannot reach you at 5:30pm places the trip elsewhere and does not call back.
This is why capacity is a revenue lever rather than an operations detail.
The costs worth attacking, in order
| Cost line | Typical share | How movable |
|---|---|---|
| Driver wages | Largest | Hard — market rate, and cutting it costs reliability |
| Vehicles & fuel | Large | Some — routing and utilization, not price |
| Dispatch | Underestimated | Very — 30-50% is documented |
| Billing & admin | Underestimated | Very — denial rate is learnable |
| Insurance | Fixed-feeling | Some — claims record drives it, over years |
A note on utilization
Most fleets asking how to earn more are asking whether to buy vehicles. Usually the answer is not yet.
Unused capacity hides in three places: no-shows, declined trips, and the gaps between broker assignments where a vehicle is on the road with nobody in it. All three are cheaper to recover than a new vehicle and none of them require capital.
Measure utilization as revenue-generating hours against vehicle-available hours before buying anything. Fleets that do this are often surprised by how much room the existing vehicles have.
What a realistic year looks like
Quarter one: measure. Missed-call rate, no-show rate, acceptance rate, real dispatch cost. Most operators do not have these and finding them out changes the priority order.
Quarter two: fix the cheap things. Answer after hours, confirm the board, call your past private riders.
Quarters three and four: work facility outreach, which takes a quarter to produce contracts and then runs for years.
Nobody moves all four levers at once, and the fleets that try usually move none.
Frequently asked questions
On broker-heavy volume, 8-15% net is the documented range for well-run operators. Fleets with a meaningful private and facility mix run higher, because those trips return 20-40% on the portion they represent.
The market is around $12.5 billion and growing at roughly 5.5% a year. But Medicare Advantage transportation benefits fell from 36% to 24% of plans between 2024 and 2026, so the growth is not evenly distributed and broker-only fleets face more risk than the headline suggests.
Whichever you can measure. In practice that is usually no-shows and after-hours answering, because both are cheap, both are fast, and both give you a baseline for judging anything else.
For a 50-trip-a-day fleet, roughly $4,000-9,000 a month against $15,000-25,000 in-house, which is where the 30-50% figure comes from. The structural benefit is that it becomes a variable cost rather than a fixed one.
Usually not first. Most fleets have unused capacity inside their existing schedule, hidden in no-shows, declined trips and gaps between broker assignments. Adding vehicles before fixing those raises cost without raising utilization.
Facility contracts take about a quarter to land and then run for years. Direct private work builds over one to three months. Neither is fast, which is why the cheap operational fixes come first.
Keep reading: maximizing NEMT profit margins, dispatch cost comparison, and startup to six figures.
