Issues No. 02 The Workup
The Operator's Brief · No. 02 · The Workup · Go-to-market · Channels

Eight and a Half Phone Calls

Three other rental companies sat in the same suburban market as Enterprise, running the same kind of operation, chasing the same claims work. Action, Agency, and Chrysler's Snappy. None of them got the channel. Enterprise ended up with nearly all of it. Working out why answers a question about your own business: which of your referrers can be won this way, which cannot, and how to tell them apart before you spend a month on it.

What the model proves

That the referral channel was won by removing the decider's work, not by cutting price; that the person who chooses the vendor is paid on rent collected, not work coordinated, so your discount never reaches her; and that the same split between who chooses and who pays runs through medical devices, auto parts, enterprise software, and freight — which is how you tell a winnable referrer from one that will waste your month.

01
If they weren't the cheapest, why did the adjusters call them?

Why being cheap did not win it

Four explanations are available for what Enterprise did, and three of them hold up to some degree. A cheaper cost base — suburban lots instead of airport counters — let Enterprise serve a low-rate segment profitably. But cheap suburban operating was available to anybody, and Action, Agency, and Snappy were all doing it in the same market. It did not get any of them the channel. National coverage is true and the wrong era: the channel was won out of a walled-off office in one body shop in St. Louis, three decades before Enterprise had a national anything. Software lock-in is probably why the position is still standing today.

The fourth explanation is the cause: they removed the adjuster's phone calls. The evidence that separates this from the others is the growth differential. Between 1998 and 2000 their insurance-replacement business grew 35 percent overall, and with the largest carriers — the ones actually wired in — it grew two to three times faster. Same cars, same rates, same branches. The one thing that differed was whether the adjuster still had to make the calls.

Friction removal won the channel. Scale and integration hold it. You can run the first. You cannot run the second — which is why the Brief tells you to expect a win with a shelf life on it.
02
What is she actually paid for?

Why she has no time

The property manager is paid a percentage of rent collected — typically 8 to 12 percent for residential — and that fee is understood to cover maintenance coordination along with rent collection, tenant communication, and reporting. On a $2,000 unit at 10 percent she earns $200 a month, and she earns the same $200 whether that unit needs nothing all year or needs you six times. Some managers add a markup on repairs, commonly around 10 percent: on a $400 plumbing job that is $40, against an hour of her time spent booking you, fielding the tenant, chasing your ETA, and checking your invoice.

So in her own accounts, every job you do is a cost. Your work generates none of the revenue she is measured on and consumes the only thing she cannot buy more of. She is not being difficult when she asks for a narrower window — she is defending the only margin she has. And the discount: if she takes a markup on vendor invoices, cutting your price cuts her fee; if she takes no markup, it lands entirely on the owner's statement and she never sees it. There is no version of the fee structure in which your discount helps the person choosing you.

03
My discount is bigger than her time is worth. Why doesn't that settle it?

Where your money is actually going

Sixteen hours of a property manager's time, at around $45 an hour once you load it, is roughly $720. The discount in the Brief's example is $3,000. You are spending four times more on the discount than the friction is costing her — which sounds like an argument for keeping the discount, and is the opposite. The two numbers do not land on the same person. The $3,000 goes to the owner's statement. The sixteen hours come out of her week. You are comparing something that helps whoever pays the bill against something that helps whoever picks the vendor, and only one of them makes the call.

Exhibit 04 · Interactive
What one referrer spends on you, in her hours rather than your dollars
Move the three sliders to your own referrer. Illustrative model, not measured data.
Hours of her year, spent on the privilege of using you
16.0
NOTES · Hours equal jobs × touches × minutes ÷ 60. The model is linear in all three inputs, so there is no threshold to find; the minutes are the number you are guessing at, and the three phone calls in the Play are what turn it into a number you can act on.
You just sized it in her hours

Get the next issue's model before anyone else.

One case, one number you believe that turns out wrong, the model and the play — free, the first Sunday, in your inbox before it goes anywhere else.

No spam, no teasers. One click to leave.
04
Is this just a car-rental story?

Where else this shape shows up

The split between who chooses and who pays is not a quirk of car rental. It is a structure, and once you can see it you will start noticing which industries have quietly organized themselves around it.

Exhibit 05
The chooser absorbs the hassle and sees none of the money
Five channels with the same shape as Enterprise's — and the axis each one actually competes on.
Medical devicessurgeon selects, hospital paysa rep in the room, not list price
Auto-parts distributionmechanic picks, owner payshow fast the part reaches the bay
Enterprise softwarethe team chooses, finance signsease of adoption, not price
Freightdispatcher picks, shipper payswhether her afternoon goes smoothly
Specialist referralsclinic refers, client paysclean notes returned fast
NOTES · In every one, the winner is whoever costs the chooser the least of their own day. If you buy businesses, revenue arriving through deciders who do not pay is stickier than it looks — and more concentrated on individual relationships than a seller will represent. Ask which named individuals route the work, and what happens when one of them moves.
05
What goes wrong when someone actually runs it?

Four ways this fails in a real shop

Techs mark jobs finished hours late and the promise breaks in public — the most common failure, covered at length in the Play. You remove the wrong touch: the three phone calls in the Play are the whole defense, and operators skip them because they believe they already know; the touch she names is often not the one you would have guessed. She moves and the relationship was personal: property-management turnover is high and everything here attaches to an individual, so get the arrangement into her company's process, not her habits. Volume arrives and service slips: share moves, the new jobs get slower response than the old ones, and it looks like success for a quarter — watch response time on that referrer's jobs specifically, not on your average.

06
What would prove this wrong?

What would change our mind

If referrers lead with price when asked. Our published prediction is that fewer than one in three will; reader replies settle it and we will print the count either way. If the growth differential has a simpler explanation — the largest carriers were also the fastest-consolidating and may have concentrated on the biggest vendor regardless of the software; we cannot rule that out. If removing a touch does not move your position on a referrer's list — this is what the Play rests on and where we have the least evidence. Readers running the play are the only source that will settle it.

Sources
Enterprise's 1994 system, the Geico Direct and Safeco request, the 8.5-calls figure and the 1998–2000 growth rates — CIO magazine reporting Enterprise's own calculation, as summarized in Auto Rental News, 2004. The 8.5 is Enterprise's own number and has never been independently checked.
The founding, the seventeen cars, and the body-shop office — company histories drawing on Enterprise material.
The 2004 share, the position of Action, Agency, and Snappy, and the spread of adjuster connectivity to small independents — Auto Rental News, 2004, citing industry estimates.
Property-management fee structures, markup conventions, and coordination fees — current industry fee guides, 2025–2026.
Marked as estimate: the forty jobs, six touches, four minutes, and the $45 loaded hourly rate are illustrative. Your own numbers replace them.
Know someone buying a referral with a discount nobody noticed?
Part 03 · The Play
You have the number. Here is the move — one referrer, four moves, about a month, and roughly six hours of your time.
Read the Play →

Axiom Research · The Operator's Brief · No. 02 · The Workup

Saved. Check your inbox.