Run your own missed-call numbers · 30 seconds

Find the jobs your shop is quietly losing.

A two-week diagnostic for HVAC contractors. I go through your call logs, quotes and ServiceTitan data, talk to your team, and hand you a ranked list of what it’s costing you, with the math shown. Then I fix the part worth fixing.

Book a 20-minute call
See a sample report
  • Fixed fee, quoted up front
  • Read-only access to your data
  • Fee credited if you build with me
Rang out after 5pm 4,100 a yearQuote never chased 296 a yearSame customer, 3 records

Example: missed calls alone

$272,800a year

79 missed × 31% × 52% × $412 × 52 weeks

I work inside the tools you already run

The problem

Busy every day. So why don’t the numbers show it?

The money slips out between the steps: the phone, the dispatch board, the truck, the invoice. In most shops it hides in four places.

  • Calls ring out after fiveAnd at lunch, when the CSRs step away. Most of those callers never get a call back.
  • Quotes go quietSent from the tablet, then nobody owns them. The customer buys from whoever calls back.
  • One customer, three recordsThe CSR can’t see who’s calling. Membership renewals go to an old address.
  • It all lives in one headWhich tech can fix what, who needs a callback. When that person is out, the shop guesses.

Run your numbers

What are missed calls costing you?

Move the sliders. It’s the same formula I use in every diagnostic, with every input out in the open.

79
31%
52%
$412

Revenue walking out the door

$272,800a year

About $103,700 of it is gross profit, at a 38% margin.

79 missed × 31% new jobs × 52% close × $412 × 52 weeks

Get my real number

Starts with the worked example from the sample report. Your real inputs come from your call log and ServiceTitan, and that is what the diagnostic measures.

How it works

Two weeks to find it. Then fix what pays.

No platform to buy and no six-month project. You’ll know what’s leaking, what it costs, and what to fix first.

  1. Diagnose2 weeks · fixed feeOne hour with you, short interviews with four to eight of your people, and your real call logs and job data. I trace records by hand to see exactly where work breaks.
  2. Build4 to 8 weeks · fixed scopeI build the top fix only, at a price agreed before I start. It’s judged against the baseline measured in the diagnostic.
  3. RunOptional · monthlyTuning, then the next fix on the list. Month to month, and you can stop any time.

The diagnostic fee is credited toward the build if you go ahead within 30 days.

See the diagnostic day by day

Why AI projects fail

Most AI projects fail for a boring reason: nobody looked at the work first.

A chatbot bolted onto messy data doesn’t fix a broken handoff between your phone and your dispatch board. It just puts a nicer face on it. So I start with the work. The tool comes last, if at all.

Read the full argument

“The most common reason AI projects fail: people misunderstood the problem the project was meant to solve.”

RAND, 2024. Interviews with 65 data scientists and engineers.

“95 percent of the organizations studied were getting zero measurable return from generative AI.”

MIT NANDA, 2025. The GenAI Divide.

Sample report, page 3: the summary, with four findings and a table of five leaks totalling about 552,000 dollars a year in expected revenue. Some figures are blacked out.
Sample report, page 8: the missed-call leak, with the formula, where each input came from, and conservative, expected and upside figures. Some figures are blacked out.
17-page sample report

What you get

A report you can check line by line

One written report, walked through on a call. Every dollar figure shows its inputs and where they came from, in three versions: conservative, expected and upside.

  • The real process map, next to the one on paper
  • Every leak priced a year, with the arithmetic
  • The top three fixes, ranked by value and effort
  • What not to automate, including what needs no AI
  • A baseline, so any work after is judged fairly
Get the sample report

Fit

Built for shops big enough to leak.

A good fit

  • Residential HVAC or home service, about 10 to 60 people
  • ServiceTitan, Housecall Pro, Jobber or similar in place
  • You suspect calls, quotes or hours are slipping
  • You’ll let me talk to the people who run the phones

Not a fit

  • Fewer than about eight employees
  • No field service software yet
  • You want a chatbot installed by Friday
  • Your staff are off limits

Nothing worth fixing? Then no build. If the diagnostic finds nothing worth the money, the report says so and I won’t try to sell you anything. If you do build, the diagnostic fee is credited.

Book a 20-minute call

Questions

Straight answers

The things owners ask before we talk. All ten questions.

Will you need to talk to my staff?

Yes. Twenty to thirty minutes each with four to eight people, scheduled around their day. The data shows what happened. Your people know why.

Do I need ServiceTitan?

No. Housecall Pro, Jobber, FieldEdge and Successware work too. You do need some field service software, because the job history lives there.

What if you find nothing worth fixing?

Then the report says so, and I won’t sell you a build. You still get the process map and a measured baseline.

What happens to my data?

Read-only access, a confidentiality agreement, never used to train AI, and deleted within 30 days of the readout unless you ask me to keep it for a build.

How much does it cost?

The diagnostic is a fixed fee set by the size of your operation. I quote it on the first call, before you commit to anything.

Next step

Twenty minutes. You’ll know if there’s money on the table.

  • You tell me what is going on in the shop.
  • I tell you whether a diagnostic makes sense.
  • No pitch deck. If it is not a fit, I will say so.

Email: [email protected]
Or send a message · Monday to Friday, 8am to 6pm Eastern

Pick a time for a 20-minute call. The calendar shows your local time.

Open the booking calendar