First call resolution (FCR) is the percentage of calls where the caller's issue is completely handled during that one call: no callback, no follow-up email, no "someone will get back to you." Most organizations that measure it properly land between 70% and 79%, and SQM Group's benchmark research puts the average closer to 71%; the same research finds a one-point gain in FCR tends to move customer satisfaction by roughly a point. Contact centres invented the metric, but you do not need one to use it. A five-person clinic, law office or plumbing company can measure it from a week of call notes, and in a small team, every avoidable repeat call is a much bigger slice of somebody's day.
What "resolved on the first call" actually means
A call counts as resolved only when the caller's stated problem is fully closed out before they hang up: the question answered, the appointment booked, the invoice issue fixed, and the same person does not contact you again about the same thing within your tracking window. A promised callback does not count. "Let me check and text you" does not count. Neither does a voicemail they had to leave twice before reaching anyone.
You will also see first contact resolution, the channel-agnostic version that counts email, text and chat alongside phone. Wikipedia's overview of the metric covers the distinctions. For a small business, the phone version is usually the one worth tracking first, because the phone is where urgent, revenue-carrying requests arrive.
How to calculate FCR without a call centre
The formula is simple: calls resolved on first contact, divided by total eligible calls, times 100. If you handled 60 real customer calls last week and 45 were fully wrapped up in one conversation, your FCR is 75%.
The real work is in two definitions you should write down before you count anything:
- Eligible calls. Count genuine customer calls only. Exclude wrong numbers, hang-ups under about 30 seconds and telemarketers. That noise skews the number in both directions.
- Repeat window. Pick a window (7 days is the common standard), write it down, and keep it fixed. If the same person calls back about the same issue inside that window, the first call was not resolved, whatever the notes say.
For a small team, the mechanics are a call log: date, caller, reason, resolved yes or no. Your phone system's call history (or the summaries from an answering service) makes this faster, but a notebook by the phone genuinely works. There are also two ways to score each call: internal, where you or your staff judge whether it was resolved, and external, where you ask the customer, usually within a day or two. Internal scoring flatters you: as Zendesk notes, customer-reported measurement is considered more accurate precisely because customers do not always agree the problem was solved. You do not need formal surveys at small volume. Asking "did we get that sorted for you in one go?" at the next visit captures most of the signal.
What counts as a good rate?
The 70–79% band is where most measured operations sit. Businesses with simple, predictable calls (booking, hours, price ranges, order status) legitimately score higher. Businesses whose calls are diagnostic, like a strange noise in a furnace or a legal question that needs a file review, legitimately need a second contact more often, and no amount of phone technique changes that.
If your number comes back above 90%, be suspicious before you celebrate. It usually means your eligibility rules exclude too much, not that you are outperforming every published benchmark. Report the numerator and denominator alongside the percentage: "45 of 60 calls" is honest in a way "75%" on its own is not.
Why repeat calls cost a small business more than a big one
In a contact centre, a repeat call is a line in a spreadsheet. In a six-person business, it is your afternoon. Every callback is staff time spent re-explaining, a job interrupted, and a customer who is a little more irritated than the last time. Research published by Qualtrics frames repeat contacts as a direct erosion of trust rather than a neutral inconvenience. A caller who has to phone three times to book one appointment has had plenty of time to call your competitor once.
And the worst repeat contact is the one that starts as no contact at all. An unanswered call almost always becomes a follow-up: either the caller tries again later, already annoyed, or they do not try again at all. If your line goes to voicemail through lunch, after 5 p.m. and all weekend, your true first-call resolution, measured from the caller's side, is lower than any log you keep will show.
Four ways the number lies
FCR is one of the easiest metrics to accidentally inflate. Watch for these patterns, even in a two-person operation:
- A shrunken repeat window. Cutting the window from 7 days to 24 hours makes the number look better without reducing a single repeat call. Most people call back within the week, not the day.
- Convenient exclusions. Leaving "complicated" calls out of the denominator boosts the rate while hiding exactly the problems you should be fixing.
- Speed over durability. Rushing callers off the line drives call length down and callbacks up. If calls are getting shorter while repeat contacts climb, that is the signature.
- Shifting definitions. If what counts as an eligible call changes every quarter, no trend you draw from the numbers can be trusted.
Raising your rate with a small team
You do not need a quality-assurance department. The fixes that move the number in a small business are mostly about preparation:
- Let whoever answers actually finish the call. If the person picking up can book the appointment, quote standard prices and answer the top questions, most calls close in one touch. If everything needs "the owner will call you back," your rate has a ceiling.
- Write down the answers to your top five repeat questions (hours, pricing, directions, what to bring, how cancellations work) and keep them beside the phone. Consistency prevents the second call that corrects the first.
- Confirm bookings before hanging up, then again in writing. A time repeated back on the call and confirmed by text or email is a booking that does not generate a "was that Tuesday or Thursday?" callback. Clinics fighting no-shows can go further with reminder calls and texts.
- Answer the next question before it is asked. The caller booking a furnace repair will want to know the service-call fee and whether they need to be home. Covering it now prevents next week's call.
- Fix the calls that ring out. Extend real coverage (a person, or an answering service that can actually resolve things) across lunch, evenings and weekends, because that is when most repeat contacts are created.
If you are comparing phone tools of any kind to help with this, judge them on what closes calls rather than on features:
| What to look for in any answering tool | Why it matters for first-call resolution |
|---|---|
| Around-the-clock, multilingual answering | An after-hours call handled tonight is far less likely to become three follow-up attempts tomorrow |
| Booking and rescheduling during the call | Scheduling requests close in one conversation instead of joining a callback queue |
| Memory of returning callers | Whoever handles the next call starts with context instead of making the caller repeat everything |
| Transfer to a person, with backup numbers | Urgent callers reach a human instead of getting stuck in an automated loop |
| Recordings and a daily summary | You can see which call types keep coming back, and fix the cause instead of the symptom |
Where AlmaTalk fits
AlmaTalk is an AI receptionist built for Canadian small businesses, and it works on the layer of first-call resolution small teams lose first: the calls that go unanswered. It answers around the clock in 30+ languages with automatic language detection (Canadian English and Quebecois French are the core pair), and it can book (or reschedule) appointments during the call into Google Calendar, Outlook or Cal.com, so a routine scheduling request is resolved before the caller hangs up rather than entering a callback queue. It remembers returning callers, follows up by SMS and email, and when someone needs a person it transfers to a human using ranked backup numbers. You get call recordings and a daily summary, which is exactly the raw material an honest FCR count needs. Across a recent 90-day window, answered calls averaged 80 seconds over 2,639 calls.
Plans are simple: Starter at $97.97 CAD for 150 minutes a month, Pro at $247.97 CAD for 400 minutes, billed by the second with no setup fee and no per-call fee. The fastest way to judge it is the same way you would judge a receptionist: call the live demo line at 1-888-775-9770 and see whether your own most common call would have been resolved first time.
Frequently asked questions
What is a good first call resolution rate for a small business?
Most organizations that measure it carefully land between 70% and 79%, and SQM Group's benchmark average is around 71%. Businesses with simple, predictable calls (bookings, hours, prices) can score higher, while diagnostic work legitimately needs second contacts. If your number is above 90%, check your counting rules before celebrating: it usually means too many calls were excluded.
How do I measure first call resolution without call-centre software?
Keep a simple call log: date, caller, reason, and whether it was fully resolved on that call. Exclude wrong numbers, telemarketers and hang-ups under 30 seconds, then check whether the same person called back about the same issue within 7 days. Divide resolved-first-time calls by total eligible calls. A week of honest logging gives you a usable baseline.
What is the difference between first call resolution and first contact resolution?
First call resolution applies specifically to phone calls, while first contact resolution is channel-agnostic and also covers email, chat and text. For most small businesses the phone version is the one to track first, because urgent, revenue-carrying requests (new bookings, emergencies, cancellations) still arrive by phone.
Do missed calls count against first call resolution?
They should. An unanswered call almost always becomes a follow-up contact: the caller either phones again, already frustrated, or takes their business elsewhere. If your line is unattended at lunch, after hours and on weekends, your real resolution rate, as the caller experiences it, is lower than your log shows. Track missed calls alongside FCR, not instead of it.
Can an AI receptionist improve first call resolution?
For the routine layer, yes. An AI receptionist like AlmaTalk answers 24/7, handles common questions from a knowledge base about your business, and books or reschedules appointments during the call, so those requests close in one contact. Calls that genuinely need a person are transferred to a human with ranked backup numbers, and recordings plus a daily summary show you which call types keep coming back.
