What Is First Call Resolution? Definition and Measurement

First call resolution explained: definition, how FCR is measured, why it matters, what lowers it, how to improve it and FCR vs repeat-contact rate.

What is first call resolution?

First call resolution, or FCR, is the share of customer contacts that are fully resolved during the first interaction, with no transfer, callback, escalation or repeat contact about the same issue within a defined window. It is measured as resolved first contacts divided by total first contacts, and it is the metric most closely tied to customer effort and cost.

Customers do not call to have a pleasant conversation; they call to get something fixed, answered or changed. Whether that happened the first time they asked is the single most direct measure of whether the contact center did its job. First call resolution captures that outcome in one number, which is why it appears in nearly every contact center scorecard and nearly every outsourcing contract. It is also one of the easiest metrics to measure badly, because the definition of "resolved" and the definition of "first" both hide decisions that change the result. This guide defines the metric, describes how it is measured and the choices inside that measurement, explains why it matters, lists what lowers it, sets out how to improve it and distinguishes it from the related repeat-contact rate. The shorter metric page is at first call resolution, and the full metrics set at customer service metrics.

How is first call resolution measured?

First call resolution is measured by taking the number of contacts resolved in a single interaction and dividing it by the total number of first contacts in the same period, then expressing the result as a proportion. The arithmetic is simple; the work is in deciding what counts in each part of the fraction. Each program should write those decisions down and report against the same definition every month.

There are three common ways to decide whether a contact was resolved:

• Agent disposition. The agent marks the contact as resolved or unresolved at wrap-up. Fast and universal, but biased toward optimism, because agents naturally believe they solved the problem.
• Repeat-contact detection. The system checks whether the same customer contacted again about the same issue within a set window, often several days, and counts the first contact as unresolved if so. More objective, but it depends on reliable customer matching across channels and on how "same issue" is defined.
• Customer confirmation. A post-contact survey asks the customer whether the issue was resolved. The most honest measure, but response rates are partial and the respondents are not a random sample.

Mature programs use more than one method and reconcile them, for example reporting agent-dispositioned FCR alongside a repeat-contact-adjusted FCR and a survey-confirmed FCR. The gaps between the three are themselves informative. The denominator also needs a decision: whether transfers, callbacks scheduled by the agent and contacts that could not have been resolved on first contact by design, such as an order that requires a supervisor's approval, are counted as first contacts, excluded, or counted as failures. Whatever the choice, it should be fixed in the contract so that neither party can move the number by changing the definition.

Why does first call resolution matter?

First call resolution matters because every unresolved contact generates at least one more contact, which costs the business a second handling, costs the customer a second effort and lowers satisfaction more than almost any other single failure. Improving FCR therefore reduces total contact volume, lifts satisfaction and loyalty measures and reduces the load on supervisors and escalation teams at the same time.

There is also a quality dimension. A high FCR usually means agents have the knowledge, the tools and the authority to act, which are the same conditions that make agents stay. A low FCR usually means the opposite: agents who have to transfer, defer or promise a callback because they cannot see the information or make the decision. For an outsourcing buyer, FCR is the clearest indicator of whether the provider's agents have actually been enabled to do the job or are simply answering the phone. It connects directly to the satisfaction measures in NPS vs CSAT vs CES, and customer effort in particular tends to track inversely with it.

What lowers first call resolution?

First call resolution falls when agents lack the information, the system access or the authority to finish the job, when routing sends the customer to the wrong skill, when policies force a handoff, or when the issue itself is one that cannot be closed in a single step. Most of the causes are structural rather than individual, which is why coaching agents on FCR without fixing the structure rarely moves the number.

The usual culprits are a knowledge base that is incomplete, out of date or hard to search; agents who can see an account but cannot change it; approval thresholds set so low that routine adjustments need a supervisor; skills-based routing that is misconfigured or too coarse, so calls reach agents trained on a different product; cold transfers that lose context and force the customer to start over; back office processes that the agent cannot see into, such as a refund that takes days to appear; scripts that end the call before confirming the issue is closed; and handle time targets so tight that agents close calls before the resolution is complete. Each of these is visible in a well-run quality program, which is why FCR analysis and quality assurance should be read together.

How do you improve first call resolution?

First call resolution improves when you identify the specific reasons contacts go unresolved, fix the knowledge, access, authority and routing gaps behind the most frequent reasons, confirm resolution before ending each interaction and measure the result against a stable definition. Improvement comes from the structure around the agent at least as much as from the agent.

A practical sequence looks like this. Start by categorizing unresolved contacts by root cause for a month, using repeat-contact analysis and agent dispositions, so you know whether the problem is knowledge, access, authority, routing or process. Then fix the largest categories in order: give agents the account actions they are repeatedly transferring for; raise approval thresholds to a level that covers routine cases; rewrite or add the knowledge articles that agents search for and fail to find, following the approach in building a customer service knowledge base; tune routing so the first agent has the right skill, as described in call routing and skills-based routing; and replace cold transfers with warm transfers that carry the context across, the difference explained in warm transfer vs cold transfer. Add a closing step in which the agent summarizes what was done and asks whether anything else is needed. Finally, make sure handle time targets leave room for resolution; the relationship between the two is covered in average handle time.

For issues that genuinely cannot be closed on the first contact, define them as such, set a next-step commitment the agent can make, and measure them separately so they do not mask the resolvable cases. A well-defined escalation path, as in customer service escalation management, is part of a high-FCR operation rather than its opposite.

How is FCR different from repeat-contact rate?

First call resolution counts the proportion of first contacts that were resolved, while repeat-contact rate counts the proportion of all contacts that are follow-ups to an earlier contact about the same issue; the two look at the same failure from different ends and are not simple mirror images of each other. A single unresolved first contact can generate two, three or more repeat contacts, so repeat-contact rate can rise sharply while FCR falls only slightly, and a program that reduces the number of follow-ups per unresolved case will improve repeat-contact rate without changing FCR at all.

Both are useful. FCR is the better outcome measure for agent enablement and customer effort. Repeat-contact rate is the better measure of total wasted volume and therefore of cost, and it does not depend on agent dispositions. A related measure, often called the resolution rate or the "resolved within one contact" measure in digital channels, applies the same logic to chat and email, where the concept of a "call" does not quite fit but the question of whether the customer had to come back is identical. Reporting FCR and repeat-contact rate side by side, with the same customer-matching logic and the same window, gives a far more honest picture than either alone.

How should FCR appear in an outsourcing contract?

FCR should appear in an outsourcing contract with a written definition of a resolved contact and a first contact, the measurement method, the window for repeat-contact detection, the exclusions, the reporting cadence and a target set from a measured baseline rather than from a general industry claim. A target the provider did not help set and cannot influence will be gamed or ignored; a target built from the first few months of measured performance and raised in agreed steps will be worked toward.

The contract should also say who owns the structural levers. If the client controls the knowledge base, the system permissions and the approval thresholds, the provider cannot be held solely accountable for FCR; the service level agreement should name the dependencies on each side. Pairing FCR with a quality score and a customer satisfaction measure prevents the easiest form of gaming, in which agents mark everything resolved. The general structure of such agreements is covered in call center service level agreement. If you want to see how FCR would be defined, measured and improved for your contact types, request a proposal and describe your channels, systems and the issues customers most often call about.