Key takeaways
- First call resolution is the share of first contacts that need no follow-up call from the customer about the same issue.
- Contact-center benchmarks put the average around 71% and world class at 80% or more[1]; measure your own office before comparing.
- You can calculate it from a two-week call log: group calls by number, find repeats within seven days, and read why they called back.
- Most repeat calls in a plumbing office come from missing arrival updates, callbacks that never landed and quotes that were promised but not sent.
The second call is the expensive one
Pull last week's call log and count the numbers that show up twice. Most offices are surprised. The second call is rarely a new job; it is “where's my plumber?”, “I never got the quote”, or “someone said they'd call me back”. Each one takes a CSR away from a new booking, and each one tells the customer the first call did not do its job.
Contact centers track this as first call resolution, or FCR. SQM Group, a firm that benchmarks call centers, puts the industry average at 71%, which means 29% of customers have to call back about the same thing[1]. It also reports that each one-point gain in FCR goes with a one-point gain in customer satisfaction and a one-point drop in operating cost[2]. Those figures come from large contact centers across many industries, not plumbing shops. The idea carries over cleanly; the exact numbers should come from your own log.
71%
Industry benchmark average[1]
70-79%
Range SQM calls good[1]
80%+
Range SQM calls world class[1]
29%
Customers who call back at a 71% rate[1]
Cross-industry call center benchmarks; a plumbing office's own number is the one to manage.
Decide what resolved means on a plumbing call
A booking call is resolved when the caller hangs up with what they needed and has no reason to call again about it. Being booked is not the same thing. A job booked into the wrong slot, with the wrong address or with no idea when the tech will arrive generates its own second call. Write the definition down for each common call reason so everyone scores calls the same way.
| Call reason | Resolved on the first call when | Not resolved when |
|---|---|---|
| New service request | Booked with a confirmed window, or told plainly why not and what happens next | “We'll call you back to schedule” with no time given |
| Price or fee question | The service call fee and how repair pricing works are explained | “The tech will tell you” and nothing more |
| Where is my tech? | Caller gets a real arrival estimate from someone who checked | “They should be there soon” |
| Reschedule or cancel | New slot confirmed and read back | Request noted for someone else to handle |
| Invoice or warranty question | Answered, or a named person calls back at a stated time and does | Message taken with no owner and no time |
| Estimate follow-up | Quote sent while on the phone, or a date for it given and met | “It'll be in your email” when nothing was sent |
Some calls cannot be closed on the first contact and should not count against anyone: a warranty claim that needs the manufacturer, a repair that needs a permit, or a caller who needs to check with a spouse. Mark those as “follow-up planned” so they drop out of the denominator, but only when the caller was told who would call, and when.
Measure it from the call log
Export two weeks
Every inbound call with date, time, caller number and, if your system has it, a call reason or tag. Two weeks smooths out a single bad day.
Remove noise
Take out spam, vendors, internal calls and techs calling the office. They are not customer contacts.
Group by number
Sort by caller number. The first call from a number about an issue is a first contact; any further call from that number within seven days is a candidate repeat.
Read the repeats
Listen to or read each candidate repeat. If it is the same issue, it counts. If it is a new job, it is a new first contact.
Calculate
FCR equals first contacts with no repeat, divided by all first contacts, leaving out the follow-up-planned calls.
Example: in two weeks an office logs 640 inbound calls. After removing 70 spam, vendor and internal calls, 570 remain, of which 460 are first contacts. Reading the repeats shows that 92 of those first contacts led to at least one call back about the same issue, some twice, for 110 repeat calls in all. FCR is 368 ÷ 460, or 80%. The 110 repeat calls are the real prize: almost one call in five in that log was a customer asking about something the office had already touched.
Seven days suits most plumbing calls; a shop that books two weeks out may need fourteen. The window matters less than using the same one every time. If your phone system tags calls by outcome, the CSR call scorecard and the booking rate in raising your CSR booking rate can be read from the same export.
Where repeat calls come from
When offices read their repeats, the same handful of causes come up again and again. Very few are about rude CSRs. Most come from promises the office made but could not see through.
- No arrival update. The window was given, then the day slipped and nobody said so. Telling customers when a tech is running late removes most of these calls.
- Callbacks that never landed. The office rang back from a number the customer did not know, got no answer and left no message.
- Quotes promised, not sent. “I'll email it this afternoon” is a repeat call waiting to happen if the afternoon gets busy.
- Vague answers on price. A caller who heard nothing concrete about fees calls back to ask again, or calls a competitor.
- Messages with no owner. A note for “the owner” or “billing” with no name and no time to call back.
- Wrong details at booking. A mistyped address or a callback number nobody read back sends the tech hunting and the customer calling.
Callbacks deserve a closer look because the customer usually did nothing wrong. In Pew's survey, 19% of US adults said they generally answer calls from unknown numbers, 67% said they don't answer but check voicemail, and 14% ignore voicemail too[3]. An office that calls back without leaving a specific message is betting on the 19%. A voicemail that names the business, the job and a direct number turns most of the other 67% into a callback instead of a repeat inbound call. Callback speed covers timing; the point here is the message.
- Generally answer 19%
- Don't answer, check voicemail 67%
- Ignore the call and the voicemail 14%
Source: Pew Research Center, 2020[3]
Show the data
| Item | Value |
|---|---|
| Generally answer | 19% |
| Don't answer, check voicemail | 67% |
| Ignore the call and the voicemail | 14% |
Fixes that cut repeat calls
- End every booking call with a read-back: date, window, address, problem and callback number. It takes twenty seconds and catches most typing errors.
- Tell the caller what happens next and when: “You'll get a call from the tech when he's on his way” is only true if it happens every time.
- Send quotes while the customer is still on the phone, or give a time and put it on someone's task list.
- Give CSRs the answers that cause the most callbacks: the service call fee, warranty basics, how financing works, what areas you cover.
- Give every message an owner and a callback time, and check them at lunch and at close.
- When you call back, leave a voicemail that names the business, the reason and a direct number.
Techs drive resolution too. The BLS lists communication as a key quality for plumbers precisely because they deal with customers regularly[4], and a tech who explains what he found, what happens next and when the office will call saves the next three phone calls. A clean call-to-dispatch handoff means the tech arrives knowing what the customer was told.
| Input | Example |
|---|---|
| Repeat calls a day | 11 |
| Minutes per repeat call | 4 |
| Share you could prevent | 50% |
| CSR hourly wage | $22.22 |
| Calls a month | 239 |
| Staff hours on these calls | 16 |
| Hours handed off | 8 |
| Monthly value of that staff time | $177 |
How it is worked out: Calls a day × 21.7 working days × minutes per call ÷ 60 × share handled = hours; hours × hourly cost.
Assumptions
- Hourly wage of $22.22: the May 2025 median for customer service representatives at plumbing, heating and air-conditioning contractors, before benefits.[5]
- Eleven repeat calls a day matches the 110 repeats over ten working days in the example above; four minutes a call and half of them preventable are set for this example.
Example only. The bigger cost of a repeat call is the customer who stopped calling and went elsewhere.
Watch it hardest in busy weeks
FCR is not steady through the year. It drops exactly when the phones are busiest: the morning after a hard freeze, the Monday after a holiday weekend, the week a tech is out. Arrival windows slip, callbacks pile up, and every slipped promise comes back as another inbound call that makes the queue longer still. Customer service work in general bunches into busy periods that can include evenings, weekends and holidays[6], and a plumbing office feels that more sharply than most.
Measure FCR separately for your three or four busiest days each month and compare them with a normal week. If the gap is wide, the fix is usually not more people on the phones but fewer reasons to call: one person whose only job on a surge day is to send arrival updates and return messages before customers chase them. Example: if half of a surge day's repeat calls are “where's my plumber?”, a status call to every customer whose window has moved, made before ten o'clock, takes those calls off the queue before they are dialed.
Not the same as first-time fix or booking rate
Three measures get confused. First call resolution is about the phone: did the call settle what the caller needed? First-time fix rate is about the truck: was the repair finished on the first visit? Booking rate is about sales: of the callers who could book, how many did? A high booking rate with a low FCR usually means jobs are booked fast and then generate status calls. Track all three and read them together.
Individual customers who call over and over are a separate problem with its own playbook in customers who keep calling. FCR is about the system that produces repeat calls, not about any one caller.
When more than one person answers
Repeat calls multiply when the second call reaches someone who does not know about the first. The fix is a shared record: notes in the customer file, call summaries everyone can see, and a habit of checking before asking the customer to start over. Recognizing repeat customers shows how a screen pop helps.
Calls answered outside the office count too. If an SI Employee covers overflow or nights, include its calls in the same log and the same FCR count. It reads the job details back before the call ends, and each summary arrives by email with the recording and transcript, so whoever picks up the next call can answer without asking the customer to repeat everything (what it does). A message it takes for the office is resolved only if the office then calls back when it said it would.
In the commercial contractor example, most of what goes wrong is routing and paperwork, not plumbing: a missing PO number or nobody at the door means another call. That is first call resolution in a commercial accent, and the same log-reading exercise finds it.
Frequently asked questions
What is a good first call resolution rate for a plumbing office?
There is no published plumbing benchmark. Contact-center benchmarks put the average around 71% and world class at 80% or above[1]. Measure your own rate for two weeks and aim to raise it, rather than chasing someone else's number.
Does leaving a voicemail count as resolving a call?
Only if the voicemail gives the customer everything they needed and they do not call back. Most people check voicemail from unknown numbers rather than answer[3], so a specific message often does close the loop.
How long a window should we use to count repeat calls?
Seven days suits most plumbing offices. Use fourteen if you book far ahead. What matters is using the same window every time you measure.
Should CSRs be scored on first call resolution?
Use it to find system problems first, such as missing arrival updates or quotes that were never sent. Scoring individuals on it before fixing those causes punishes people for things they cannot control.
Is first call resolution the same as booking rate?
No. Booking rate measures how many callers who could book did. First call resolution measures how many callers had no reason to call back about the same issue.
References
SQM Group (2026). First Call Resolution (FCR) Metric: Operating Philosophy (opens in a new tab) ↑
SQM Group (2026). First Call Resolution (FCR) Metric: Operating Philosophy (satisfaction and cost) (opens in a new tab) ↑
Pew Research Center (2020). Most Americans don't answer cellphone calls from unknown numbers (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). Occupational Outlook Handbook: Plumbers, Pipefitters, and Steamfitters (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (Public Data API) (2025). Occupational Employment and Wage Statistics, May 2025: Customer Service Representatives at plumbing, heating and air-conditioning contractors, hourly median wage (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). Occupational Outlook Handbook: Customer Service Representatives (opens in a new tab) ↑