Northeast Freight Time one clock, four readings

Measuring Your Own Operation

Everything else in this section is about reading other people's numbers. This is the one that produces yours, and it is the only data that describes your operation. For another workplace-measurement reference, ways time trackers can be manipulated.

Four weeks. No purchase. The output is four figures per facility and a cause list, which is enough to reprice a lane, redesign a dock schedule or settle an argument that has been running for a year. For complementary transport data and research, see Logistics Management.

Week one: capture only

Six columns, fifteen seconds per visit. Date and load, facility, gate in, gate out, mode, and a three-word cause note when the visit ran long.

Do not analyse anything. The temptation after four days is to draw conclusions from four days. Resist it; the variance in freight is large enough that a week of data will tell you a confident and wrong story.

And record every visit. Including the twenty-minute ones. The fast visits are what make the slow ones credible, to you as much as to anyone else.

Week two: add the split

Once the habit holds, add check-in time — the moment the driver was entered into the facility's system.

This single field separates the yard segments and answers the question that decides where the problem is: is the wait before or after check-in?

Before means a queue outside the gate, which is a capacity or appointment-density issue. After means a door assignment problem. Different owners, different fixes, and no amount of dwell averaging distinguishes them.

Weeks three and four: keep going

Nothing new. You are accumulating enough observations for a distribution.

Thirty visits at one facility is a floor for a usable median and a rough 75th percentile. A stable 95th needs more, and a quarter is better.

What to compute

Five things, and none needs anything beyond a spreadsheet.

Median gate-to-gate, per facility. The typical visit.

75th percentile. What a bad-but-normal visit looks like, and the number to schedule against.

Proportion over your free-time allowance. The only figure that predicts your invoice volume.

Gate-in to check-in, median. The invisible segment.

And the cause list, sorted by frequency. Three-word notes cluster fast, and two or three causes usually account for most of the long visits.

Do not compute the mean. It will be pulled by the tail and it describes almost no actual visit.

What to do with it

Reprice, or ask for a change. With a number rather than an impression, and stated as a method the other side can check.

Take the cause list to whoever owns the cause. At least one entry is usually fixable by scheduling rather than by spending.

Fix your free time. If a large share of normal visits breach your allowance, the allowance is wrong, not the visits.

And set a baseline. Repeat the four weeks in six months. Comparing your own periods is the only comparison where the definition is held constant — which is more than any published benchmark can offer.

What it will not tell you

Whether you are normal. No facility-level benchmark exists and none will, because the data is commercially sensitive.

Whose fault it is. It tells you where the time goes, not who should pay, and that is a contract question.

And it will not settle a dispute retroactively. Data collected from now on is evidence about now. That is a reason to start today rather than after the next argument.

The honest expectation

Most of what you find will be unsurprising and one thing will not be. The morning cluster, the one slow door, the commodity that always runs long, the day of the week nobody had noticed.

That one thing is usually worth more than the four weeks cost, and it is not findable any other way.

The short version