Last summer I decided to actually manage my tractor instead of just owning it.

It’s a 2018 John Deere 1025R — a subcompact diesel that mows, loads, and throws snow for us year-round. After eight years and 350 hours, its entire documented history was one messy note and my memory. Every homeowner with equipment knows this state. The machine works, so you don’t look too hard.

I looked.

What the audit found

The last fully documented service was three years and 170 hours ago.

The fuel filter was thirty-one months into a twelve-month replacement interval. Greasing had “happened regularly” — says the guy who never once logged it. And the hydraulic filter, which the service schedule wants changed at 50 hours and every 200 after, had, as far as any record could prove, never been changed. Eight years.

The machine ran fine. That is exactly how deferred maintenance likes to look, right up until it doesn’t.

Best of all: I had never read the serial number plate. On this platform the serial number gates which parts actually fit — there’s a mid-run engineering break — so I had been one confident online order away from a box of wrong filters and an afternoon I’d have enjoyed considerably less.

None of this is unusual. That’s rather the point. I have spent my working life building systems so that organizations don’t operate on somebody’s memory, and I had been operating a diesel engine on mine.

The honest-data rule

Rebuilding the maintenance record, I hit the tempting lie that every backfilled system invites: write down what “must have” happened as though it were fact.

I know I greased it. Roughly. Probably in the spring. Call it April. And once you’ve written April, it’s a date, and dates look like facts, and in two years you will believe it.

The rule I settled on instead: work I can’t date gets a question mark and shows an amber “verify” flag — forever, until a real dated entry replaces it. The log refuses to pretend it knows what it doesn’t.

That one decision changed how the whole system feels. When the dashboard says something is current, it’s current — not “probably fine.” The amber flags are ugly and I have not been in a hurry to clear them, which is itself information: eight years of habit doesn’t get reconstructed in a weekend, and a record that admits the gap is more useful than one that papers over it.

I stole this rule from how I insisted reported numbers work at my day job. It turns out it works just as well in a barn.

A deferral is a decision — if you write it down

The overdue fuel filter didn’t get changed that week.

Parts, timing, season — I made the call to defer it to the end of August. Which, stated plainly, is the same thing I had been doing for thirty-one months.

The difference is one line in a log. The deferral is recorded, with a date and a reason, and the overdue flag stays up where I can see it. Procrastination and planned deferral are the same inaction. What separates them is a timestamp and an owner.

I think this is the most portable idea in the whole project. Most organizations I’ve worked in are perfectly good at deciding to do things and perfectly bad at deciding to not-do things. The not-doing happens anyway; it just happens by default, unwritten, unowned, and impossible to revisit because nobody ever chose it. Writing it down converts a drift into a decision, and a decision can be argued with.

Then I built the cockpit

Because I can’t leave anything alone, the tractor now has a full digital cockpit.

A 3D model where every implement is clickable. Service items with color-coded status. A diagnose mode that walks symptom checklists. And — the part I’m unreasonably proud of — the complete factory parts catalog, 445 sections of exploded diagrams, mapped to the machine.

Built on an assumed serial range, both sides of the parts break captured, because the plate is still unread. It’s on the list. The system won’t let me forget. That’s the point of the system.

I want to be honest about the ratio here: the audit was worth doing and took an afternoon. The cockpit took considerably longer than an afternoon and was, by any reasonable accounting, unnecessary. I built it because I enjoyed building it. Some of the best things I’ve made started as an excuse to keep working on something, and pretending otherwise would be its own kind of dishonest data.

But it did teach me something I didn’t expect. Once the parts catalog was mapped, I stopped guessing at repairs. Not because the catalog fixed anything — because having the diagram in front of me changed what I was willing to attempt. That’s the thing I’m actually trying to teach my sons, showing up in my own garage: this is knowable, and I’m allowed to try. Turns out it works on adults.

The scorecard

The project graded out at “done — maintain only,” which in my world is the highest state a system can reach. Not “improving.” Not “in progress.” Done, and now it just needs feeding.

The snowblower’s exact model is still unconfirmed until I read a decal. The machine is the same machine it was last spring.

The difference is that now I know what I know about it — and, more useful, I know exactly what I don’t.

■