Home / Guides / First call resolution
Metrics · 7 min read
First call resolution: how to measure it without fooling yourself
First call resolution is the most useful metric in customer service and the most commonly mismeasured. Almost every organisation reporting a healthy FCR is measuring something narrower than what the name implies.
What FCR is supposed to mean
The customer made contact with a problem. The problem was resolved. They did not have to come back.
That is the whole idea, and its value is that it correlates with almost everything else you care about: satisfaction, cost to serve, retention. A customer who has to chase you is a customer who is being taught that chasing is necessary.
The four ways the number gets flattered
In practice FCR is usually measured in one of these ways, and each one is systematically generous:
- Asking the agent. A tick box at wrap-up: was this resolved? The person answering has both an optimism bias and, in most schemes, an incentive.
- Counting tickets, not contacts. If a second call about the same matter opens a new ticket, both tickets can close as first-contact resolved. The customer contacted you twice.
- A short window. Measuring repeat contact within 24 or 48 hours misses the customer who waited a week for a promised callback before ringing back. That is precisely the customer most at risk.
- Matching on the phone number. The same person calling from a mobile instead of the office line reads as a different customer.
Each of these is defensible on its own terms. Together they can turn a real FCR in the sixties into a reported figure in the nineties.
The version worth measuring
A number you can act on has four properties:
- Measured from the customer's side. Did this person contact us again about this matter? Not: did we think we finished it.
- Matched on the matter, not the ticket. Two contacts about one unresolved booking are one failure, however many records were created.
- A window that matches your promises. If you tell customers "someone will call you within five business days", a 48-hour window cannot detect your most common failure.
- Blind to channel. A customer who rings, then emails, then rings again has contacted you three times.
Why transcripts make this possible
The reason FCR is usually measured badly is that measuring it well is hard with the data most businesses have. Call detail records know the number, the time and the duration. They do not know what the call was about, so they cannot tell a customer ringing back about an unresolved repair from the same customer ringing about something new.
Transcripts change that. When the content of each call is available, contacts can be linked by subject rather than by phone number - the same customer, the same matter, across days and across different agents who each believed they were handling something fresh.
This is also the point at which FCR stops being a scorecard number and starts being diagnostic. You are no longer asking "what is our FCR" but "which matters generate repeat contact, and why" - and that question has an actionable answer.
What good looks like
Be suspicious of a very high number. An FCR above ninety per cent usually means the measurement is generous rather than the service is exceptional. Genuine, customer-side FCR in a business with any operational complexity typically sits meaningfully lower, and that is not a failing - it is the honest baseline you need before you can improve anything.
What matters more than the level is the breakdown. A single FCR figure tells you almost nothing. The same figure split by matter type, by system involved and by supplier tells you where to spend, and usually reveals that a large share of repeat contact traces back to a small number of causes.
Watch what happens when you publish it
FCR responds to being measured in ways that are not always improvements. If agents are scored on it, the reliable ways to raise the number are to record the matter as resolved, to avoid creating the second record, or to keep the customer on the line longer than the problem requires.
Measuring from the customer's side rather than from the agent's disposition removes most of that pressure, because there is nothing on the agent's screen to adjust. It is another reason to derive the number from what actually happened on the calls rather than from what was typed afterwards.
Common questions
What is a good first call resolution rate?
Be careful with benchmarks, because most published figures are measured generously. An FCR above ninety per cent usually indicates a permissive measurement rather than exceptional service. The more useful comparison is your own figure over time, measured consistently from the customer's side.
How do you measure FCR accurately?
Measure repeat contact from the customer's perspective: did this person contact you again about the same matter, within a window that matches the promises you make, across any channel, matched on subject rather than on ticket or phone number.
Why is FCR hard to measure from call records?
Call detail records contain the number, time and duration but not the subject. Without knowing what each call was about, you cannot distinguish a customer ringing back about an unresolved issue from the same customer ringing about something new. Transcripts make that distinction possible.
See what a full review of your calls turns up
WiseSentry reads every recorded call and raises only the ones where your business let the customer down. Talk to us about a review of your own call history.