Part 02 · Build vs. buy · Residential

Utilization is the whole game

A real Tuesday, minute by minute. You are not buying surveys. You are buying minutes, and only about a third of them are survey minutes.

InteractiveResidential8 min read

Part 01 ended on an assumption worth pulling apart: that the surveyor is busy all day. Almost nobody is, and the shortfall does not show up anywhere in a P&L. It shows up as a cost per survey that keeps drifting up while everyone insists nothing changed.

Here is a Tuesday. Two jobs booked, a third slot open because a homeowner moved to Thursday.

TimeWhat is happeningMin
7:00Shop. Load ladder, charge tablet, get the day.25
7:25Drive to first job.32
7:57Arrive. Homeowner conversation, dog, gate.14
8:11Roof, attic, panel, shading.55
9:06Pack up, notes, confirm nothing was missed.21
9:27Drive to second job.38
10:05Nobody home. Call, wait, call the office.22
10:27Dead slot. The Thursday move left nothing here.46
11:13Lunch.30
11:43Drive to third job.29
12:12Arrive, access, setup.12
12:24Roof, attic, panel, shading.58
13:22Pack up and notes.18
13:40Find signal, upload photos, fix two that failed.35
14:15Drive home.45
Paid480
Actually capturing site data113

Twenty-four percent. Nothing went wrong on this Tuesday. Nobody was lazy, no truck broke down, the weather was fine. One homeowner rescheduled and one was not home, which is a normal week, and the day still returned under two hours of the work the customer is actually paying for.

The surveyor was not underworked. The day was.

Why this is the number that matters

Because every fixed cost in Part 01 divides by it. The truck costs the same on a two-survey day as on a four-survey day. So does the insurance, the tablet, the salary. Utilization is the denominator under all of them at once, which is why a ten point swing in booked density moves cost per survey more than any wage negotiation you will ever have.

It is also the number that is hardest to fix from the inside. The levers are real but slow: tighter geographic clustering, a scheduler who batches by zip rather than by request order, confirmation calls that actually reduce no-shows, a capture standard that stops sending people back. Every one of those is a quarter of work, not a week.

Run your own Tuesday

Interactive model

Where the paid day actually goes

Set the shape of a normal day. The tiles show what you are really buying.

min
jobs
min
min
min
min
$/day
Capture utilization
23%
of paid minutes
Idle or unbooked
70
minutes with no job attached
Cost per capture hour
$256
what the productive hour costs
Cost per survey
$239
at this day's output

The reason outsourcing changes the arithmetic

Not because somebody else's surveyor is cheaper per hour. Usually they are not. It is because the idle minutes stop being yours. A dead slot on a partner's Tuesday is a cost they absorb across a national book of work, and the density they can build across many customers in one metro is density a single installer cannot build from their own pipeline alone.

That is the entire structural argument, and it is worth stating plainly because it is often oversold. Outsourcing converts a fixed cost into a variable one. That is good when your volume moves and bad when it does not. If your volume is flat, high and geographically tight, an in-house crew running at genuine density is hard to beat, and you should keep it.

The honest caveat

Utilization measured this way flatters nobody, including us. Any partner quoting you a per-survey price is running the same equation on their side, and their answer depends on how much other work they have within thirty minutes of your job. Ask about density in your metro specifically. It is a fair question and the answer is diagnostic.

Ask us about density in your metro

It is the question that decides whether outsourcing is cheaper for you specifically, and it has a real answer.

Book a working session