First call resolution: measuring it honestly and actually improving it

First call resolution (FCR) is the proportion of customer contacts fully resolved in the first interaction, with no callback, no follow-up ticket and no second attempt by the customer. It correlates with satisfaction better than almost any other support metric, because a repeat contact means the customer spent their time twice on the same problem. It is also one of the easiest metrics to report favourably while the underlying experience stays unchanged, which is why the definition matters more than the number.

Defining first call resolution so the number means something

• Count a contact resolved only if the same customer does not return about the same issue within a set window — seven days is common.
• Match across channels: a call followed by an email about the same problem is not a resolution.
• Exclude nothing quietly. If you remove sales calls or known outages, publish the exclusions with the figure.
• Measure from the customer's side — "resolved" means their problem ended, not that the agent closed the ticket.
• Report alongside average handle time, because pushing one without watching the other simply moves the cost.

Never set first call resolution as an individual agent target on its own. It rewards closing tickets prematurely and discourages the escalations some problems genuinely need. Measure it at team and journey level.

What actually moves the number

1. Find the repeat drivers: group returning contacts by root cause rather than by agent, and fix the top three.
2. Give agents authority to decide — refund limits, credit thresholds, exception approvals — because most repeats are waiting for permission.
3. Put the answer where the agent is: a knowledge base article linked from the ticket beats an intranet search.
4. Fix upstream defects that generate contacts; support volume is usually a symptom of something earlier in the product or billing process.
5. Route accurately the first time; a transfer is a strong predictor of a repeat contact.
6. Close the loop on causes with the teams that own them, and track whether the driver disappears.

Measuring any of this assumes the conversations live somewhere countable. Customer support software supplies that: a queue with owners and states, response targets you can report against, and a history the next person can read.

Answering at the first attempt gets easier when the common answers already exist. The five most-used saved replies also describe the five things customers cannot work out alone, which is a product backlog rather than a support achievement.

The trade-off with handle time

Teams under pressure on average handle time will end conversations early; teams pushed only on resolution will over-invest in simple cases. The useful pairing is first call resolution with contact rate per customer: if resolution rises while contacts per customer fall, the improvement is real. If resolution rises and total contacts stay flat, the definition is probably leaking.

Ettex Chat helps on the operational side: conversation history per customer across sessions, so an agent can see whether this is a repeat before answering, and canned responses that keep the correct answer one click away. Repeat contacts become visible in the history rather than hiding behind a closed ticket.