Everybody in the trades has a feeling about missed calls. The feeling is usually "probably fine, they call back." A feeling is hard to act on. A number on a sheet of paper is not.
Here is a worksheet you can finish tonight with your phone, your invoices, and a calculator. Nothing here comes from a survey or an industry report. It comes from your own books, which is the only version of this number worth trusting.
Step one: count the calls you never picked up
Every phone keeps a record. On an iPhone it is Recents, and the red entries are the ones you missed. On Android it is call history. If your shop line runs through a carrier app or a VoIP dashboard, you can usually pull thirty days and export it.
Print or screenshot the last thirty days. Mark every call where nobody talked to a human.
Now take the noise out. Your crew calling from the van is not a lost job. Neither is the supply house, the parts store, the recruiter, or the same number ringing you four times in ten minutes because they were determined to reach you and did. Spam comes out too.
What is left is unique phone numbers that tried to reach your shop and never got a person. Write that count down. Call it M.
Step two: get your average ticket from invoices, not memory
Pull your last ninety days of invoices. Add up the dollars, divide by the number of invoices. That is your average ticket.
Memory will lie to you here, because you remember the panel upgrade and the full exterior repaint, not the twelve service calls in between.
If your work splits into two obvious buckets, run two averages. Auto repair might split oil changes and diagnostics from brakes and suspension. A painter might split one bedroom from a whole house. A garage door tech might split a broken torsion spring from a new opener with a keypad. Then use whichever mix actually shows up on the phone.
Write it down. Call it T.
Step three: your close rate comes from the calls you did answer
Take one week of answered calls from people who were not already customers. How many of those turned into scheduled work?
You do not need software for this. Tally marks on a sticky note by the phone for a week will do it. Two columns: talked to a new caller, booked the work.
Divide bookings by calls. If you booked six out of ten, that is 0.6. Call it C.
Be honest here rather than generous. This number is the closest thing you have to proof, because it is what happens when you actually get to talk.
Step four: the first multiplication
M times C times T. That is one month of first visits you did not run.
Say your log shows 18 unique callers you never spoke to, you book about half the new callers you do speak to, and your average ticket is $240. That is 18 times 0.5 times 240, or $2,160 for the month. Those three numbers are placeholders. Use yours.
Write the result at the top of the page. That is the floor, not the ceiling, and the next two steps explain why.
Step five: add the second visit
A new customer is rarely one invoice.
Pull your customer list for the last twelve months. Count how many customers have more than one invoice on it. That share tells you how often a first job becomes a relationship.
For a rough lifetime figure, divide total invoices for the year by the number of unique customers who generated them. If 400 invoices came from 250 customers, the average customer bought 1.6 times. Multiply your step four number by that.
The shape of this changes by trade, and you know your own shape better than anyone:
Pool service and house cleaning are not selling one visit, they are selling a route slot that either stays full all season or does not. A barber is selling a chair every three or four weeks for years. An electrician who does a panel upgrade in March is often the one who gets the call about ceiling fans in the summer and the hot tub circuit the following spring. A plumber who clears a line once becomes the name written inside the utility room door.
Step six: leave referrals at zero
You cannot count what you never saw. Leave the referral line blank and put a zero in it.
That feels like giving up value, and it is. It also makes your total defensible, which matters when you are deciding whether to spend money to fix the problem. A number you can walk anyone through beats a bigger number you had to guess at.
Step seven: find out when the phone rings
Go back to the missed call list and sort it by hour and by day.
Most shops find clumps rather than an even spread. Lunch. The window between four and six when people get off work and finally deal with the thing they noticed on Saturday. Saturday morning itself. Around St. Louis, the first hard freeze fills a plumber's line with frozen hose bibs and a no heat call by suppertime, the first stretch of ninety degree days does the same for anyone who touches air conditioning, and one bad storm takes out power and garage doors across three counties at once.
Then add the reason column, because it is usually the same short list. Hands in a drain. Under a dash with the lift up. On a ladder with a brush. Mid haircut. Running a compressor loud enough that you would not hear the ring anyway. Driving on I-44 with a trailer behind you.
Those clumps tell you what kind of fix you need. A phone problem that shows up every weekday at 4:30 is a different problem from one that shows up only during the first cold week of November (more on that in who is actually calling at 7pm).
Do it tonight
One sheet of paper, seven lines: missed unique callers, average ticket, close rate, first visit total, repeat multiple, adjusted total, and the three hours of the week when it happens most.
Tape it inside a cabinet door. Run it again in thirty days and see whether the number moved.