What is first contact resolution: FCR explained
First contact resolution (FCR) is the share of customer contacts resolved during the first interaction, with no transfer, no callback and no follow-up ticket inside a defined window after the contact closes. It is the headline quality metric in most support operations, and the one most often measured badly.
It is called first call resolution when the channel is voice only. First contact resolution covers phone, chat, email and messaging in one number, which is what most multichannel teams report.
How is first contact resolution calculated?
FCR equals contacts resolved on the first interaction, divided by total contacts in the period, times 100.
Three settings decide what that number actually means, and all three have to be written down before anyone reports it:
- The reopen window. How long after closing does a repeat contact still count against the original? Seven days is common. Drop it to 48 hours and the rate climbs several points without one extra issue being solved.
- The unit of counting. Contacts or issues? A customer who calls three times about one problem is one issue and three contacts. Count contacts and the failure lands twice in the denominator, flattering the percentage.
- The source of truth. Agent-marked resolution, system-detected repeat contact, or a post-contact survey. The three produce materially different rates on identical work, and only the last asks the person who knows.
What counts as resolved
Agent-marked FCR measures whether the agent believed the issue was closed. That is a useful proxy and a poor truth. The three definitions in use, in ascending order of honesty:
- Agent disposition. The agent selects "resolved" at wrap-up. Cheap to collect, and optimistic by about the margin you would expect from any self-assessment.
- Repeat-contact detection. The system looks for another contact from the same customer, on the same topic, inside the window. Harder to fool, and it catches the customer who called back rather than reopening the ticket.
- Surveyed resolution. The customer is asked whether the issue was fully handled. The most accurate, the most expensive, and reliably the lowest of the three.
Transfers are the other definitional fight. A cold transfer to a second queue is not first contact resolution by any reading, and most teams arguing that a warm one counts are arguing to protect the number.
Why FCR and average handle time pull against each other
These are the two numbers support teams are measured on, and inside a single contact they are in direct conflict. Every minute an agent spends confirming the fix, checking the adjacent thing about to break, or explaining the workaround properly is a minute added to average handle time.
Cut the contact short and AHT improves this week. The repeat arrives next week, in a different reporting period, usually with a different agent who starts from zero.
The arithmetic is not close. Call Centre Helper publishes an industry standard AHT of 6 minutes 3 seconds, drawn from 190,702 entries into its Erlang Calculator. Rushing a contact might save 40 seconds. The repeat it causes costs another six minutes, plus queue time, plus the customer's patience. Trading 40 seconds for 360 is a bad deal made thousands of times a week.
The research points the same way. The study of more than 75,000 customer service interactions behind Stop Trying to Delight Your Customers in Harvard Business Review concluded that service teams should focus on problem solving rather than speed, and head off the downstream follow-up instead of closing fast.
First contact resolution: a worked example
A support team handles 12,400 contacts in a month and marks 9,300 resolved at first contact, reporting 75 percent. The repeat-contact check disagrees: 1,050 of those customers came back inside seven days on the same issue. Real FCR is 66.5 percent. That 8.5 point gap is 1,050 contacts the operation paid to handle twice, none of them in the staffing forecast, because the dashboard said they never happened.
How FCR gets gamed
- Marking it resolved and hoping. The most common distortion, and rarely deliberate. Agents close what they believe they fixed.
- Shortening the reopen window. A 48-hour window beats a seven-day one every time. Nothing about the service changed.
- Reclassifying the repeat. New ticket, new subject, no link to the original. The second contact enters the month as fresh demand.
- Attaching it to agent pay. Dispositions get generous, repeats get recategorised, and you lose the data while keeping the problem.
- Reading the blended average. One product line or fault family usually carries most of the failures. Segment before concluding anything about agent skill.
How FCR differs from first-time fix rate
They measure the same idea at two different points in the service chain. FCR is the contact-centre metric: did the interaction close the issue without a second contact? First-time fix rate is the field equivalent: did the engineer who attended finish the job without a return visit?
They run in sequence. A contact that fails FCR often becomes a dispatch, and a dispatch that fails first-time fix becomes another visit. Teams reporting both usually find one root cause under each: the person who had to decide could not see what they were deciding about. It is also why raw ticket deflection substitutes for neither.
That visibility gap is where guided capture fits. Venta Capture, a product of VentaVid, sends the customer a link, walks them through recording the views and answers the agent needs, and returns a structured visual submission the agent can open during the contact rather than asking someone to describe a fault in words. On its customer support page Venta Capture reports 23 percent of customer problems solved without dispatching a technician, a vendor-reported figure rather than an industry benchmark. It does not diagnose anything. It gives the agent something to look at while the customer is still on the line, which is the only moment first contact resolution is available. The same shift one step further down the chain is covered in remote diagnostics in field service.