First Call Resolution (FCR): The percentage of customer contacts resolved completely on the first interaction, requiring no callback, repeat contact, or internal escalation to close the same issue. A high FCR rate means customers are not calling back about the same problem, which cuts contact volume, lowers cost, and raises satisfaction simultaneously.
What Does First Call Resolution Actually Mean in Practice?
FCR means the agent had the knowledge, authority, and tools to fix the issue in a single contact. That sounds simple. In practice it is harder than it looks. An agent may close a ticket in the system while the customer’s problem is still open. A caller may not ring back but might email, chat, or complain on social media. Those silent failures do not show up as repeat contacts, but they are still FCR failures.
The practical definition I use: the issue is resolved when the customer does not need to contact the company again about the same matter within a defined window, typically five to seven business days. That window needs to be agreed before measurement starts, because vendors define it differently. I have seen vendors report FCR above 85% using a 24-hour window and the same operation sitting at 68% on a seven-day window. That gap matters.
How Is FCR Calculated?
The standard formula is straightforward:
FCR (%) = (Contacts resolved on first interaction / Total contacts) x 100
Example: if an outsourced support team handles 10,000 calls in a month and 7,800 are resolved without the customer calling back, FCR = 78%.
Measurement methods vary. The two most common are:
| Method | How it works | Strength | Weakness |
|---|---|---|---|
| Internal / system-based | Flag repeat contacts from the same customer within the window | Scalable, consistent | Misses cross-channel repeats (call then email) |
| Post-call survey | Ask the customer “Was your issue fully resolved?” | Captures customer perception | Sampling bias, survey fatigue |
| Combined | System flags + survey sample | Most accurate | More setup required |
I always recommend the combined approach for outsourced environments. Relying purely on system data inflates FCR because it misses channel-switching. Relying purely on surveys under-represents dissatisfied customers who skip the survey.
What Is a Good FCR Rate for an Outsourced Contact Center?
From operating experience, a reasonable indicative benchmark for a general inbound customer service operation is 70% to 80% FCR. That is not a hard standard. The right target depends on the complexity of the product, the authority given to agents, and the quality of the knowledge base.
Context matters a lot:
| Contact Type | Indicative FCR Range |
|---|---|
| Simple transactional (order status, basic billing) | 80% to 90%+ |
| General customer service (mixed) | 70% to 80% |
| Technical support, L1 to L2 | 60% to 75% |
| Complex or regulated queries (insurance, healthcare) | 50% to 70% |
These are indicative ranges from operational experience, not published benchmarks. Your product complexity, your agent empowerment policy, and your escalation paths will shift these numbers.
If a vendor quotes you 90%+ FCR for a complex product without qualification, I would push back hard. Either their measurement window is short, they are defining resolution loosely, or they are only reporting the easy contacts.
Why FCR Matters When Evaluating a BPO Provider
FCR is one of the few metrics that shows the cost and quality side of an operation at the same time. A low FCR means customers contact you multiple times for the same issue. Each repeat contact is a cost. It is also a dissatisfaction signal. Repeat callers are angrier, take longer to handle, and are more likely to churn.
I tell buyers: if a vendor has a low cost-per-call but a poor FCR, the real cost per resolved issue is higher than it looks. A vendor charging $12 per hour with 65% FCR may cost more per resolution than one charging $15 per hour with 80% FCR, because every third or fourth contact is rework.
For buyers evaluating call center outsourcing providers, FCR is the metric I would anchor SLA conversations around, alongside CSAT. It ties agent knowledge, training quality, tool access, and empowerment into a single number.
Country and delivery model affect FCR too. Offshore teams in the Philippines or India can achieve strong FCR on well-documented processes. Where FCR tends to suffer offshore is on high-context, ambiguous queries that require judgment calls and where the knowledge base has gaps. The fix is almost never “hire better agents” — it is usually better documentation, clearer escalation authority, and a stronger knowledge management process.
How FCR Relates to Other Metrics
FCR does not sit alone. It connects to:
- Average Handle Time (AHT): Agents who rush calls to hit AHT targets often generate repeat contacts. Chasing AHT at the expense of FCR is a common mistake in outsourced environments where productivity bonuses reward speed over resolution.
- CSAT / NPS: FCR and customer satisfaction are strongly correlated. Customers who had their issue resolved in one contact almost always score higher. A drop in FCR usually signals a coming drop in CSAT within the same reporting period.
- Contact volume: Every FCR failure creates at least one extra inbound contact. Improving FCR from 70% to 80% on a 50,000-contact-per-month operation removes roughly 5,000 repeat contacts. That volume reduction is real cost savings.
- Escalation rate: High escalation rates suppress FCR. If agents cannot resolve issues because they lack authority or tools, FCR falls no matter how skilled the agents are.
If you are comparing vendors, ask for FCR alongside AHT and CSAT. The three together tell you much more than any single metric alone.
Before signing with any outsourced contact center, ask the vendor how they measure FCR, what their definition of resolution is, and what window they use. Then ask what actions they took in the last quarter to improve it. A vendor who gives you a number without a story around it probably does not manage to it. Get quotes from vendors who report on FCR as a live operational metric at /get-quotes/.