What does one missed call cost an electrical contractor?

Your hands are inside a panel and the phone's in the truck. It rings, it stops, and when you get back down there's a number you don't recognize and no voicemail. You shrug it off. Probably a price shopper.

Here's the short version. A missed call costs you the average value of a call to your shop, and for an electrical contractor that average is a lot higher than the service call you're picturing. Your phone carries troubleshooting work, small installs, service and panel upgrades and the occasional generator, all on one line, all sounding the same for the first ten seconds. You can't know which one rang out, so the honest price of an unanswered ring is the blended number across all of them.

Why the number in your head is too low

Ask an electrician what a job is worth and you'll usually hear something in the low hundreds. That's an honest answer about the most common call, and it's the wrong input for this.

An electrical phone carries at least four different businesses. Troubleshooting and repair: a dead circuit, a breaker that won't hold, half the kitchen out. Small install: fixtures, fans, a dedicated circuit, most EV chargers. Service and panel upgrades. And the occasional standby generator or whole-house rewire.

Those aren't variations on one job. They're different price bands, different close rates, and a different buyer on the other end. The one you think of first is the one that rings most often, which is also the cheapest one you do.

So the error isn't in anybody's formula. It's in the ticket people plug into it.

The one line, and what it isn't

Here's the calculation:

Value of one call = the sum, across your job types, of (number of calls of that type × your average invoice for that type × the share of them you book), divided by your total calls

Be clear about what this is, because it's easy to oversell. It isn't a cleverer formula than average ticket times close rate. It's the same quantity, total expected revenue over total calls, and further down you'll see the two agree to the dollar.

What the split buys you is honest inputs. "Average ticket times close rate" only works if you use your true blended booked ticket and your true close rate across every call. Almost nobody has those in their head. What they have is one job type standing in for all of them. Bucketing is just the method that stops you doing that.

Splitting your calls into buckets that mean something

One rule: keep splitting until each bucket is tight enough that its average describes the jobs inside it.

That rule does real work here. "Project work" is a tempting bucket and a useless one, because a $1,200 EV charger on a panel with a spare breaker and a $16,000 standby generator with a pad, a gas line and a transfer switch are not the same business. Averaging them gives you a number that fits neither.

Split them and both averages start meaning something. Four or five buckets is usually enough for an electrical shop. If a bucket still has a job ten times the size of another job in it, split it again.

Once a bucket is tight, use the plain average inside it. People reach for a median because of skew, and a bucket you've narrowed properly doesn't have much left. A median would understate your total anyway, and you're about to multiply it by a call count.

Where the three inputs actually come from

Calls by type. If you already tag calls in your CRM, pull ninety days. If you don't, and most shops don't, you can't reconstruct it from memory. Put a notepad on the dash for thirty days instead and write one line per call: the date, what it turned out to be, and booked or not. Thirty days is short enough that you'll actually finish it.

Average invoice by type. Your invoicing, same window, one average per bucket.

The share you book, by type. This is the one almost nobody has, and it's the one that moves the answer most. It's also the reason the notepad has a booked-or-not column.

One carve-out before you start. Work that never touches the general line, the builder who has your cell, the dealer who sends you installs, belongs in its own bucket or out of the exercise entirely. Folding relationship work into a per-call average inflates a number you're about to spend against.

If you can't produce a booking rate yet, there's a published figure to anchor against. Invoca's 2026 home services report found that 38% of calls answered by a person are leads, and that 45% of those leads convert on the call, which works out to roughly 17 of every 100 answered calls converting while the caller is still on the line.

Read it as a floor, not as your booking rate. Its denominator is every answered call, junk and wrong numbers included, and it only counts what closes during the call. Yours counts real inquiries, and the quote you win on Thursday. The report also covers nine home services sub-industries, and electrical isn't one of them, so it's the neighborhood rather than your address.

A worked month for a two-truck shop

Every number below is invented so the arithmetic is visible. Replace all of them. Say 120 calls in a month:

• 78 troubleshooting and repair calls, average invoice $380, booked 75% of the time
• 30 small install calls, average $1,200, booked 50%
• 9 service and panel upgrades, average $4,200, booked 30%
• 3 generator and large project calls, average $14,000, booked 20%

Run the four lines:

• 78 × $380 × 0.75 = $22,230
• 30 × $1,200 × 0.50 = $18,000
• 9 × $4,200 × 0.30 = $11,340
• 3 × $14,000 × 0.20 = $8,400

That's $59,970 of expected work sitting inside 120 calls, or about $500 a call on these made-up inputs.

Check it the crude way and you get the same answer, which is the point made earlier. This shop books about 77 of its 120 calls, a 64% close rate, at a blended booked ticket of $781. And $781 × 0.64 is $500. Same number, because it's the same quantity.

Now compare it to the number in your head. If you'd priced a ring off the call you picture, $380 booked three times in four, you'd have said $285. The real figure is about three quarters higher, and most of that gap is the twelve calls at the top: 10% of the volume, about a third of the value.

Miss one call in ten, which is another guess you should replace, and twelve calls a month at $500 is $6,000 a month. Call it $72,000 a year on those inputs. Not because twelve generators walked out the door. Because the blended average walked out twelve times, and generators are in the blend.

What this number ignores: your capacity

Here's the objection every owner raises in the first ten seconds, and it's the right one.

Two trucks can't deliver $59,970 of work every month and also absorb another $6,000 of it. If you're booked two weeks out, the next call isn't worth $500. It's worth the difference between that job and whatever it would have displaced, which can be close to nothing.

So read it honestly: it's what a call is worth when you have room, and an upper bound when you don't. Most shops sit between the two, busy some weeks and quiet others.

The mix argument survives the capacity argument, though. If you're full and a panel upgrade rings out, you didn't lose a slot. You lost the chance to trade a $380 service call for a $4,200 one, and that trade only gets offered to a shop that knew the call came in.

Why the big calls are the expensive misses

The dollar gap is the obvious half. The other half is what happens to those calls when you don't pick up, and it splits two ways rather than one.

Some are shopping. A homeowner planning a service change is gathering two or three quotes, and the shortlist gets built out of whoever they actually reached. A callback doesn't fix that. You're not returning a call, you're asking to be added to a list that's already full.

The rest are the opposite, and they're the ones people forget. A mast torn off the house by a limb, a red-tagged meter, a panel flagged on an inspection with a closing date attached, an insurer with a deadline. Those are four-figure jobs where the caller hires the first competent shop that answers. Both kinds pay well, and both punish a missed ring.

An old measurement points at the shopping half. A study of 1.25 million sales leads at 29 B2C and 13 B2B U.S. companies, reported by Oldroyd and colleagues in Harvard Business Review, found firms making contact within an hour were nearly seven times as likely to qualify the lead, meaning to get a meaningful conversation with a decision maker, as those that tried even an hour later. That's 2011 data on online sales leads, so take the direction and leave the multiple.

And you mostly won't be told it rang. Invoca's platform data puts the share of voicemail-bound callers who leave a message at under 3%, which is covered in whether customers leave a voicemail at all.

Where owners undercount this

Four places, and the first is easily the most common.

Counting the callback as a save. You rang back at six. They booked the shop that answered at two. It shows up nowhere in your numbers as a loss, because it was never in your numbers to begin with.

Leaving spam in the denominator. If a fifth of your inbound is junk, you're understating what the real calls are worth by roughly that much. Strip it out before you divide, then treat screening as its own problem.

Assuming your misses look like your calls. They probably don't. Quote calls cluster in the evening and on Saturday, which is exactly when nobody picks up, so the calls you lose are likely worth more than your overall average, not less.

Pricing the miss at revenue when you're spending against it. Revenue is the right unit for sizing the problem. If you're using the number to make a spending decision, run the same four lines on gross profit per job so you're comparing like with like.

What actually moves it

Three things, and they aren't equally available to you.

More calls. That's marketing, it costs money up front, and it takes months to show.

A better booking rate on the calls you already get. That's training and follow-up, it's slower than it sounds, and the same Invoca report found on 55% of the home services calls Invoca scored, nobody asked the caller to buy or book at all, so there's usually room in it.

Or answering more of the calls that already ring. That one doesn't need more demand and it starts working the day you fix it. Invoca puts the share of home services callers who speak with a person at 52%, on a denominator carrying the same junk and hang-ups yours does, so read it as a marker of how common the gap is rather than as a target. How fast you get back to a new lead sits in the same place.

Run your own four lines first

Thirty days with a notepad on the dash, then the invoicing for the same window, then the arithmetic. The tagging is the work and the arithmetic is ten minutes. What you get is your number instead of a benchmark that half fits, and the general version of this math for any trade is in what a single captured call is worth.

Then go look at what your phone does between four and eight in the evening and on a Saturday. That's where the quote calls sit, and usually where nobody's picking up. If you're working on the answering side already, how electricians avoid missing service calls while they're on a job covers the setup.

Willison answers your calls 24/7, in seconds rather than rings, and it answers several at once, so a caller who lands while you're already on the line doesn't get a busy signal. It asks what kind of job it is, gets the details down, and either books it or hands you a call that's ready to quote. What it won't do is dispatch anybody, send a truck, or promise a homeowner a time.

If you'd rather hear it than read about it, press play on a real call on willisonhq.com and listen to how much of the detail it gets down. It is an HVAC call; yours runs on the same engine, built around your own services and intake questions.

Frequently asked questions

Then your inbound line is carrying a smaller share of your revenue than this exercise implies, and the honest thing is to say so in your own numbers. Keep dealer, builder and realtor work in a separate bucket and run the calculation on the general line alone. You'll get a smaller per-call figure and a truer one.

Four or five is plenty for most electrical shops, and the test isn't the count, it's the spread. If the biggest job in a bucket is more than a few times the smallest, the average you get out is describing nothing you actually sell, and the bucket needs splitting.

None of them on their own. Take a full quarter if you can, and if your year has a genuine peak and a genuine trough, run it twice and keep both numbers. The seasonal shape matters more than the annual average here, because the weeks when your phone is busiest are the weeks you're least able to answer it, and that's when the per-call figure is highest.

Some comes back, and honest math says so. Somebody who's used you before will usually try twice. A stranger collecting quotes generally won't, and since under 3% of voicemail-bound callers leave a message you rarely even find out who they were. The conservative version is to apply a recovery rate to your repeat callers and none at all to first-time ones.