Stock from what you actually fit, not what you might need

Ask a tech what should be on the van and you get the list of everything that has ever been needed. The useful list is different: the parts that come up often enough to be worth the money and the space.

Go through the last few months of completed jobs and count what went in. The pattern is usually sharper than anyone expects — a short list of items covering most calls, and a long tail of things used once.

Work out what each shortage actually costs

Not everything deserves the same treatment. A part that stops the job dead and means a second visit is worth carrying even if it is rarely used. A part you can pick up on the way, on a job that is booked for next week anyway, probably is not.

Sort the list by what happens when you do not have it, rather than by how often it is used. Those two orders are not the same, and the first one is the one that matters.

Set a minimum per item, and let the system tell you

Running out is almost never a decision. It is nobody noticing. A minimum quantity per item turns that into something the system notices for you: when the count drops under the number, the row is flagged, and at zero it is flagged harder.

Set the minimum at what you use between restocks, plus enough to cover the gap if an order is slow. For a fast-moving part on a weekly restock, that is a week's use plus a cushion.

Keep the count honest

An inventory nobody updates is worse than none, because people trust it and get caught out. The only version that survives contact with a real crew is one where adjusting it takes a couple of seconds on a phone, in the driveway, right after the part goes in.

The other half is being able to see who changed what. Not for blame — for finding out where stock is actually going, which is usually onto another van or into a job nobody logged.

Count it properly now and then

Even a well-kept system drifts. Parts get swapped between vans in a hurry, something gets used on a warranty call-back nobody wrote up, a box turns out to have been half empty.

A physical count once a quarter is enough for most shops. If the drift is consistently in one direction, that is information worth chasing down rather than just correcting.

Know what you are carrying in money

Hold a unit cost against each item and the van has a total. Most owners are surprised by it the first time — a well-stocked service van is often carrying several thousand dollars that could have been in the bank.

That number is not an argument for carrying less. It is the number that makes the trade-off a decision rather than an accident, and it is useful at the end of the year.

Review it when the work changes

A van stocked for the work you did two years ago is carrying dead weight. Equipment changes, the mix of jobs changes, a new building goes up nearby and suddenly a different fitting is the common one.

Look at it once or twice a year: what has not moved in six months, and what you have gone back to the supplier for more than twice.