Method
The Revenue Capture Layer
Between “phone rings” and “job booked” sits an operating layer no line item shows. This page is the whole method: what the layer is, how it gets measured, and how an engagement actually runs.
The framework
The Revenue Walk.
Demand you already paid for steps down through five operational leaks to what actually books. Each step of the walk carries observed incidence from the field study, and inside an engagement the same walk is rebuilt from your own call and booking data, in dollars.
The five leaks
Where the demand actually goes.
Dies at answer.
The call reaches voicemail, a dead end, or an answer-only service while paid media is live. The demand existed for exactly one phone call.
Price opacity.
The person answering cannot state what a visit costs or whether the fee credits against the work. Callers comparison-shopping hang up and keep shopping.
Never recovered.
The caller defers ("I’ll talk to my spouse") and nobody asks for an email, sends a recap, or schedules a callback. The lead exists only in the caller’s memory.
Replacement handled as a ticket.
A buyer holding a competitor’s quote is offered a service visit days out, and nobody asks to see the quote. The biggest ticket in the trade is processed like a drain call.
Attach missed.
A second problem volunteered mid-call, the exact cross-sell the platform thesis priced in, gets a second visit, a second fee, or "call someone else."
The measurement
One call, one rubric. This is what a Read scores.
Every inbound demand call crosses the same stages. A fixed rubric of observable behaviors, scored the same way on every call, every wave. The early stages run on software these platforms already bought; the rest are habits, scripts, and configuration. The exhibit below carries the July 2026 baseline results. A Read applies the same rubric to your own brands, so this is the sample output with the market’s numbers where yours would sit.
Answer
49/55Reached a live voice. Six dead ends, all at brands with live paid media that week.
Qualify
no scoreNot scored: the rubric has no clean denominator here, because what counts as a qualifying question varies by trade. Read qualitatively it was the strongest stage in the sample, though one weekend emergency still ended without a phone number taken.
Price
40/49Stated the visit fee, or free. Nine could not price a visit when asked.
Paid software ends here · workflow begins
Hold
8/49Tried to keep the caller at the moment they deferred. Six of the eight were scripted AI agents.
Capture
13/49Asked for an email. One address announced out loud as fake was accepted.
Recap
1/55A written recap received before hang-up. The best call in the sample, and the only one.
Follow-up
2 obs.Two promised callbacks observably arrived. The rest fell outside the study window and are unscored.
Route
0/55Live referrals to a sibling brand. Not one, in any phrasing.
Observed incidence, 55 calls placed as a customer to 48 PE-owned brands, July 2026. Denominators vary because six calls never reached a voice and some stages only trigger in certain scenarios. The first three stages run on software these platforms already bought. The last five are habits, scripts, and configuration.
How the numbers are labeled
Every number on this site carries a label.
Consulting numbers usually arrive naked, and you are left to guess what is holding them up. Here the label does that work, on this site and in every read-out I deliver. If a claim can’t carry one of these labels, it doesn’t ship.
- Observed
- Counted in the field: calls placed as a customer and scored on one rubric, or behavior seen directly in a client’s systems. No extrapolation.
- Measured
- Before-and-after counts from a client’s own booking system, with the counting method disclosed.
- Modeled
- Arithmetic on stated inputs: flow-through, multiples, brand counts. The inputs are published, and they are yours to argue with.
- Recommended
- A judgment call about what to do next, labeled as one.
The Bookability Baseline cites outside data, so it adds two more labels on the same principle: External benchmark for a figure found in a cited dataset, with vendor research flagged as vendor research, and Interpretation for a reading of the evidence, with the competing reading printed beside it.
Why now
The hold period is the deadline.
The flagship case counted roughly 120 extra booked jobs a month at one brand; priced at the platform’s average ticket, that is about $470K a year. At 25 to 30 percent flow-through that is roughly $120K to $140K of EBITDA, and at eight to twelve times, $1M to $1.7M of enterprise value, per brand that looks like that one. On a five-year hold, a quarter of delay is a quarter of that gone, permanently, and the fixes move at configuration speed. The arithmetic is modeled and the inputs are published; bring your own numbers.
The objection
The dashboard number and the diligence number.
You probably have a booking rate on a dashboard already. It was measured from inside: your own recorded calls, counted your own way. Invoca’s aggregate says 52% of home services callers reach a person. ServiceTitan’s platform data put the average shop’s booking rate at 42% (2022). A buyer’s diligence team will use neither. They will call your brands like a customer and count what happens. I run that call program now, before it prices your exit.
Both figures are external benchmarks published by vendors in this market, linked so you can check them and weigh the source. Neither is a ZTS number.
What happens next
Four steps, each one optional after the last.
- 01Thirty minutes
Operating review
One platform, one commercial question. You leave with where I’d look first, what to measure, and whether ZTS is the right help. If it isn’t, I say so and point you at who is.
- 02Two weeks · Fixed fee
Revenue Leak Read
Every brand shopped from outside, graded against the 48-brand baseline on a fixed rubric of observable behaviors, and three fixes ranked by payback. It ends with a decision meeting rather than a proposal: the read is complete in itself, and if we continue, the Read's fee credits in full against the Sprint.
- 03Your call
The decision
Three honest outcomes: your team runs the fix list alone; I run the highest-dollar items in a ninety-day Sprint; or the number is too small to bother, and the read-out says so.
- 04Ninety days, then standing
Sprint, then Managed
The Sprint fixes and measures one platform. The Managed tier keeps the leak map current across brands and onboards each acquisition onto the same playbook, priced per brand.
Fair questions
The questions a diligence team would ask, answered in advance.
What access do you need, and who holds the keys?
The Read needs nothing from your stack. No logins, no data room, no reporting access, nothing installed. I shop the brands from outside, as a customer, and grade what happens. Access only becomes a question in a Sprint, where changes go through your own admin under change control and every configuration is documented so your team owns it after I leave. Client data stays in your systems; excerpts leave only anonymized, and an NDA is standard.
Was the field study legal and ethical to run?
The calls were placed to published consumer lines, presenting as a homeowner asking things any caller may ask. No bookings were confirmed, no technician was dispatched, no free work was extracted, and where a rep offered to hold a slot I declined or released it on the call. Scoring was done from transcripts against one rubric, and every brand and sponsor is anonymized. The full method and ethics note is in the report appendix.
You’re one person. What happens if you get hit by a bus?
Everything ships as an artifact your team owns: the read-out, the fix list, the configuration documentation, the report definitions. The deliberate rule is that you can run the plan without me. Key-person risk on a two-week diagnostic is two weeks; on the Managed tier it’s bounded by the documentation, and you can test that claim by reading a redacted read-out before you sign anything.
Our CMO’s spend is what this makes look bad. Why would they back it?
The opposite: the study’s whole finding is that marketing did its job. The demand arrived; operations dropped it. Call-to-booked-job by brand is the number that proves the CMO’s spend is working and shows exactly where its return is being destroyed after the phone rings. In most platforms this work moves budget pressure off marketing, and onto a fix that costs configuration time.
Why is there no price on this site?
The fee is fixed and agreed before the work starts, and it scales with platform size, so a single public number would be wrong in both directions. The shape is public: two weeks, fixed fee, sized against one brand’s annual leak rather than a day rate. Ask on the first call and you leave with the figure.
Can you run this inside a diligence window, before we own the platform?
Yes, and it is a different product from the post-close Read, so I name it differently. Outside-in work cannot produce a dollar leak number: with no booking data and no close rates, a handful of calls to a target’s brands will not carry that weight, and I won’t pretend otherwise in front of your CDD provider. What it does produce, in about ten days and with no target cooperation, is whether the cross-sell machinery exists at the front line, how far intake practice spreads across the brands you are about to pay for, and the questions to force into the data room. It lands as an appendix your deal team can drop into the IC memo, alongside commercial diligence rather than instead of it.
Where do our call recordings actually go?
They stay in your systems wherever the work allows it, and I score from your existing recordings rather than pulling copies by default. Where transcription is needed, it runs on a named processor agreed with you in writing before anything moves, in the jurisdiction you require, with no training on your data. Anything I hold for the engagement is deleted at close-out on a stated date, and excerpts in any read-out are anonymized. Call recordings are consumer personal information, so your GC gets these terms up front rather than at contracting.
If we hire you, do our brands show up in the next published wave?
No. Client platforms are excluded from the public baseline during an engagement and for twelve months after it ends. The baseline exists to be evidence you can trust, which it stops being the moment a brand wonders whether it is being shopped by its own advisor.
Something unanswered? Email Zaha and get the straight version.
Start here
Two weeks and a number.
The operating review is thirty minutes and decides whether the Read is worth running.