What the National Transit Database wants, in plain language

The reporting concepts small agencies trip over — service hours versus revenue hours, unlinked passenger trips, deadhead — and how to capture them without a second data-entry job.

Most NTD pain is not caused by the reporting. It is caused by the fact that the numbers were never captured during operations, so somebody has to reconstruct a year from paper at the end of it.

The distinctions that cause the most trouble

Term	What it means	The mistake
Revenue hours	Time available to carry passengers	Including the drive from the yard
Deadhead	Movement with no passengers permitted	Not separated, inflating revenue hours
Unlinked passenger trips	Boardings — each vehicle boarded is one	Counting a transfer as a single trip
Vehicles operated in maximum service	Peak vehicles actually out	Reporting the whole fleet
Service hours	All operating hours including deadhead	Used interchangeably with revenue hours

If you take one thing from this page: capture deadhead separately from the first day. Reconstructing the split after the fact is close to impossible and it distorts every cost-per-hour figure you will ever quote.

Capture at the moment, not at the deadline

Every figure NTD asks for is a by-product of something that already happened operationally. A run either operated or it did not. A vehicle either went out or it did not. If those facts are recorded when they occur, the annual report becomes an export instead of a project.

Record the run as operated, cancelled or missed at the end of each day. One tap, not a form.
Record boardings at whatever granularity you honestly can, and be consistent about it.
Separate revenue from deadhead in the schedule itself, so the split is structural rather than remembered.
Log the reason for a cancellation. It costs five seconds and it answers the board's question in advance.

The interior of a small historic depot: a large round wall clock above wainscoting, an empty bench, and long afternoon window light across the floorboards. — **Plate 12**Built by people who assumed it would still be running now. It is.

Counting boardings when you cannot count perfectly

Almost no small agency has automatic passenger counters, and a driver operating a dial-a-ride cannot reliably tally boardings while also driving. That is a real constraint, and the answer is not to pretend otherwise.

What matters far more than precision is consistency. A count that is systematically five percent low, applied the same way by every driver on every run for three years, still produces a usable trend and a defensible cost per trip. A count that changes method whenever staff change produces nothing at all.

Write down the counting rule in one sentence, and put it where drivers see it.
Decide explicitly how a transfer, a personal care attendant and a no-show are each counted. These three account for most of the disagreement.
Have new drivers count alongside an experienced one for a week before their numbers go in.
If you change the method, record the date you changed it. A visible step in the data is fine; an invisible one is not.

Report what happened

A quiet temptation in this work is to smooth the numbers — to report the schedule rather than the service. Do not. Beyond the compliance exposure, the practical problem is that a smoothed number cannot support a request. If eleven percent of trips were missed because you are one driver short, that number is the argument for the driver.

Source: National Transit Summaries and Trends — Federal Transit Administration

Questions

We are a small rural operator. Do we even report?

Reporting obligations depend on the federal funding you receive and your size — rural Section 5311 recipients typically report through their state DOT rather than directly. Your state transit office is the right first call, and they are generally helpful.

Can software file it for us?

No, and be suspicious of anyone who says otherwise. What software can honestly do is hold the underlying counts accurately all year so that filing is transcription rather than archaeology.