FCR is the most gamed metric in support

The ticket said "Resolved" on Friday. By Tuesday it was back: same customer, same broken export, but a new ticket number and a clean timestamp, because the original thread had aged past the reopen window. On the dashboard, first-contact resolution scored one for one. In reality, the problem was never fixed, just relabelled.

That gap is why FCR deserves more suspicion than any other number in support. Not because resolving an issue on the first try is a bad goal, it's the goal, but because FCR has the widest gap of any metric between what it claims to measure and what it actually rewards. It is trivially easy to inflate, and almost every trick that inflates it pushes reps to close conversations before the customer is done.

The definition is the loophole

Start with the fact that nobody agrees on what "first," "contact," or "resolution" means. Is a follow-up email from the same customer a new contact or a continuation of the old one? Does a ticket the customer reopens count against you, or does a fresh ticket reset the clock? Is "resolution" the customer confirming they're happy, or an agent clicking a status field on the way to lunch?

Each of those choices is a dial, and every dial turns the same direction: looser definitions push the number up. The published benchmarks look reassuringly precise —

— but that figure is only comparable if two teams count the same events, and they almost never do. A team that treats every reopen as a brand-new issue and lets agents self-mark resolution can manufacture a world-class FCR without changing a single customer outcome.

Reopen windows hide the failures

The reopen window is the quietest cheat of all. Set it to 24 hours and any problem that resurfaces on day two arrives as a new ticket, a new "first contact" that will probably also resolve on first contact. Narrow windows don't improve resolution; they just move repeat contacts outside the measurement frame. The customer lives one unbroken problem. Your reporting records three tidy successes.

Speed pressure does the rest

Then there's the clock. Handle time and FCR are supposed to be independent, but in practice they're wired together. Benchmark average handle time sits at

and reps who feel that pressure learn that the fastest way to "resolve" a contact is to answer the question that was asked, not the problem underneath it. Close the ticket about the failed password reset; ignore that the account is locked because billing suspended it. FCR ticks up. The customer comes back, to a different queue, a different agent, a different ticket that also resolves on first contact.

The cost lands somewhere the FCR chart can't see. Customers are less forgiving than they used to be:

A premature close isn't a neutral event. It is one of the bad experiences that sends people to a competitor, and your FCR dashboard will glow green the whole way down.

What an honest FCR looks like

None of this means abandon the metric. It means stop treating it as self-reporting. Two habits keep it honest.

First, pair it with conversation quality you actually inspect. QA review scores exist precisely to catch the technically-closed ticket, and the industry internal quality benchmark runs high:

A resolution a reviewer wouldn't sign off on shouldn't count as one. So measure resolution from the customer's side, not the status field: ask them if it's fixed, and watch repeat-contact rate across a 7- or 14-day window regardless of ticket number, so reopens can't hide behind fresh IDs. And keep FCR next to satisfaction. When the resolution rate climbs while CSAT flatlines, you are almost certainly looking at gaming rather than progress. A number that improves alone, unaccompanied by any signal from the people it supposedly serves, is telling on itself.

Second, respect how hard resolution actually is. Only

of service issues get fully resolved in self-service, which means most of what reaches a human is genuinely unresolved and genuinely hard. A metric that makes that work look easy is not measuring the work. It is flattering it.

FCR is worth keeping. But treat a high number as a question, not an achievement: the first thing to audit, not the thing to celebrate. When it moves, ask whether resolution changed or the definition did. For the numbers behind the claims here, see the published support benchmarks; for why repeat effort is often the more honest companion signal, start with the customer effort score.