You don't need a calculator with sliders on it. You need last month's phone bill, the price of a normal job, and about ten minutes at the kitchen table. Most vendors would rather you skipped this, because the arithmetic comes out either obviously yes or obviously no, and both answers take the same ten minutes.

Four numbers. That's the whole exercise:

  • How many calls came in last month
  • How many of those nobody picked up
  • What share of answered callers turn into paying work
  • Your gross margin on an average job (margin, not the ticket price)

Minutes one to three: pull the real list, not the summary

Log into your carrier's account and find call detail or usage detail. Verizon, AT&T and T-Mobile all bury it a couple of clicks past the bill summary. If your business line is VoIP, the call log sits in the admin panel and exports as a CSV. If your number just rings a mobile handset, the Recents list only holds so much, so go to the carrier record anyway.

Quick aside, then back to the math. The phone bill is the last honest document in a small business. The books get optimistic. The pipeline in your head gets very optimistic. Your Google review page is a lie by selection: only the delighted and the furious ever bother to type. The call log has no opinion. It says a stranger rang you at 4:52 on a Tuesday, rang for 22 seconds, and stopped.

Minutes three to five: count the ones that died

Three categories: missed, sent to voicemail, and answered but under fifteen seconds. Count that third one as a loss. Twelve seconds means the caller heard a recorded greeting start up and hung up on it.

Now subtract the junk, honestly, because this is where people cheat in the flattering direction. Say the log shows 214 inbound calls and 71 that nobody handled. Knock off 22 obvious robocalls. Knock off 8 more where the same number rang three times in four minutes, which is one annoyed person and not three. Knock off 6 suppliers and your kid asking for a lift. You're left with about 35 strangers a month who wanted something and got nothing.

A 411 Locals study of 85 businesses found 62 percent of calls to small businesses go unanswered, which is a small sample and a good reason to check your own log rather than trust the headline. That invented log above misses 33 percent. Still 35 people. The Numa Small Business Phone Report puts it at 85 percent of callers who reach voicemail never calling back, which matches what anyone who has ever hunted for a plumber on a Saturday already knows. A missed call doesn't wait in a queue for you. It goes to whoever picks up next.

Minutes five to seven: what one call is actually worth

Two numbers, and please use margin. Revenue gives you a lovely figure that means nothing.

Say you run a two truck garage door outfit in Erie. Average job is 240 dollars. After the spring, the labor and the fuel you keep about 40 percent, so 96 dollars. And roughly one in four strangers who get through books something. So every genuinely new inbound call is worth about 24 dollars of profit before anybody answers it. That's your unit. Write it on the corner of the bill.

Minutes seven to nine: multiply, then be mean about it

35 missed strangers times 24 dollars is 840 dollars of margin a month walking off. Now halve it. Some of those people rang back an hour later and you caught them. Some were price shoppers who'd never have booked anything. Some will hear an AI, decide they hate it, and hang up. Call it 420 a month.

Being pessimistic on purpose is what makes the number worth anything. Any vendor calculator that asks you to enter an "estimated lift percentage" is asking you to invent the answer and then be impressed by it. Use a figure you made too small deliberately. If it still clears, you know.

Minute nine: the cost column, which is boring

Rocketship has a free tier, and paid plans start at 12 dollars a month. So 144 dollars a year, against 96 dollars of margin on one spring swap.

Break even is a job and a half. Not a month. A year.

That's why this takes ten minutes instead of an afternoon. When the cost side is smaller than one repair call, precision stops mattering. Be wrong by a factor of five in either direction and the sign doesn't change. If you're already paying an answering service by the minute, or a part timer four hours a day to sit near the phone, put those in the column too and the gap gets embarrassing.

When the honest answer is no

Sometimes the log says don't bother. Listen to it.

If 9 calls came in last month and you answered all 9, close the tab. Nothing to recover. If your work comes from three commercial accounts who email purchase orders and never ring, the phone isn't the leak. If your average ticket is 18 dollars the arithmetic gets tight, though at 12 dollars a month it usually still clears. And if a caller needs a human being inside the first ten seconds, medical triage, a crisis line, then this is the wrong tool and I won't pretend otherwise.

Everyone else: the figure on the corner of your phone bill is the price of the current arrangement, where the phone rings while you're up a ladder with a torsion spring in both hands.

The question that's left

Once break even is a job and a half a year, price stops being the interesting part. What's left is whether the thing holds up at 9:40 on a Sunday night. Does it answer on your own number, so the caller doesn't see an area code they don't recognize and hang up? Does it know you charge 240 for a spring swap and don't touch commercial roll ups? Does the appointment land in the Google Calendar that's already open on your laptop? Does it take a name and a callback number, so even a bad call leaves you something to work with? And does it hand the caller to you when they want a person?

All of that is testable in an afternoon. Get a number, ring it yourself at 9:40 on a Sunday, and listen to what a stranger would have heard.