How to calculate average dwell time for machines on a dealer lot

Dwell time, in this context, is how long a piece of equipment sits at a customer's site once it's out in the field, from the day it shows up on a job to the day it gets traded, sold, or scrapped. That's different from the retail-lot meaning of the term, new units parked at the dealership waiting for a buyer. If you're planning replacement campaigns across a territory, this number tells you more than almost anything else a dealer's service records can give you.
Here's the basic math, then why it breaks down once you're covering more than a few sites.
The manual calculation
For a single machine, dwell time is just arrival date subtracted from departure date. Average dwell time for a site is the sum of those individual dwell periods divided by the number of machines that passed through, over whatever window you're measuring (a year, a sales cycle, whatever matches your campaign planning horizon).
To do this by hand you need two things nobody keeps consistently: a logged arrival date for each unit, and a confirmed departure date. Most dealers have neither. Service records tell you when a machine came in for work, not when it first landed on a customer's yard. Warranty registration tells you when it was sold, which might be years before it shows up at the site you're watching. So the practical version looks like this:
- Pick the sites in your territory worth tracking, usually the larger accounts or the ones overdue for a refresh.
- Walk or drive each yard and log what's there: make, model, rough age if you can tell from the unit, and the date of your visit.
- Revisit on some fixed interval, ideally no more than a quarter apart, or dwell estimates get too coarse to act on.
- Match what you saw last visit against what you see now. Machines that are gone have a dwell time you can close out. New arrivals start their clock. Whatever's still sitting there adds another interval to its running total.
- Average the closed-out dwell times for the site, and keep a separate eye on anything still running past two or three intervals. Those are your oldest fleet.
That's the whole method. The math itself is simple; step 2 and step 3 are what break down at scale. One rep can walk a dozen yards a quarter if the territory is tight and nothing else is on the calendar. Across forty or fifty accounts spread over a few hundred miles, that visit cadence falls apart by the second quarter, and once it does, your dwell numbers are really just guesses dressed up as data.
Why the interval matters more than the precision
A common mistake is trying to get dwell time down to the day, when what the campaign planning needs is the bucket: is this fleet turning over inside 18 months, or is a chunk of it still the same iron you saw three visits ago? That second question is the one that tells you a site is overdue. Precision to the week doesn't change which sites get a call this quarter. Consistent visit intervals do, because they're what let you compare "still there" against a known baseline instead of eyeballing age off a hunch.
The whole job comes down to a reliable pass-to-pass comparison, run on a schedule you can sustain across the whole territory, not just the handful of yards closest to the office.
A quarterly, site-by-site turnover map is built around exactly that comparison. Instead of a rep logging arrivals and departures by hand, each site in the territory gets checked on the same quarterly cadence, and machines seen last pass are matched against machines seen this pass, so you get an arrival and departure timeline per location without anyone driving the route.
What to do with the average once you have it
Once you've got a dwell average per site, or even just a rough count of what's been sitting since three passes back, sort your territory by it. The sites with the oldest standing fleet and the slowest turnover are your first calls for a replacement push. The sites turning over fast don't need the same attention this quarter. It's a prioritization list before it's anything else, and it only works if the underlying visits actually happened on schedule.
If logging arrivals and departures by hand across your territory has stalled out by the second quarter, it might be worth seeing what a dropped, dated turnover map looks like for your sites instead.