Blog Viewer

Call Center Metrics and KPIs to Measure Performance and Productivity


There is a great deal of debate about what metrics call centers should measure now, especially with COVID-19 and the accompanying shift to work-at-home agentsstaffing shortages, and persistent supply chain issues straining customers’ patience.

Despite changes in technology, new contact channels, and omnichannel communication, some call centers are still "partying like it's 1998," using outdated metrics and operating as a cost center. Times have changed, and where necessary, call center and BPO metrics must change with them.

Here, we answer the following eight questions regarding metrics that call centers should use today:

  1. What are call center metrics, and how have they evolved?
  2. What metrics should call centers and BPOs focus on?
  3. What key metrics should call centers measure for voice, chat, email, and SMS?
  4. Are there unique metrics for work-at-home versus in-center?
  5. Should we keep measuring AHT?
  6. Should we focus on resolving the customer's issue and aim for high Overall Satisfaction (OSAT)?
  7. Do new measurements like employee satisfaction scores (internal NPS) matter?
  8. How do BPOs measure client sentiment?

What are call center metrics, and how have they evolved?

Call centers derive metrics at the agent and center levels to gauge the overall effectiveness of agent teams regarding performance, productivity, and other customer satisfaction-related activities. Operations managers monitor key performance indicators (KPIs) to assess agent efficiency. Call center executives track strategic metrics and KPIs to make a business case for good customer service on behalf of their clients.

While the list of KPIs is endless, standard metrics that have stood the test of time include Quality Assurance (QA), Customer Satisfaction (CSAT), First Call Resolution (FCR), After Call Work (ACW), Service Level, and Occupancy.

However, metrics have changed along with the industry. In the past, call centers primarily interacted with customers by telephone to resolve inquiries quickly and at the lowest cost.

Customers dialed a toll-free number and talked with an agent to address their issue. Due to the phone-based nature of the system, customers could not contact businesses through their preferred channel or at their convenience.

Call centers measured customer satisfaction based on the speed at which an agent answered the phone and responded, regardless of whether the request was resolved. This hurried approach frequently left customers feeling dissatisfied.

With the shift to multi-or omnichannel contact centers, today's emphasis is on customer experience. That means operational metrics like Average Speed of Answer (ASA), Average Handle Time (AHT), and FCR prove to focus on controlling costs instead of ensuring overall customer satisfaction. Now, customer-centric measures, such as CSAT, Customer Effort Score (CES), and Customer Lifetime Value (LTV), stand alongside internal metrics to measure contact center success.

What key metrics should call centers measure for voice, chat, email, and SMS?

Incorporating email, messaging apps, and other digital channels into your customer service and support mix lets you track more metrics than in the past—and it’s smart to monitor the number of interactions received per channel for staffing purposes.

That said, every customer service and support channel will have the basics of Quality Assurance, Service Level, CSAT, FCR, and ACW. Also, all will measure AHT as a forecasting indicator to determine the resources needed to handle volume at specific intervals.

Chat and SMS are the new kids on the call center customer service block and have a set of metrics unique to them. Some of the more common include:

  • Average Resolution Time (ART) - The average time it takes for an agent to close a conversation. It's a derivative of AHT.
  • Chat to Conversion Rate - A lead-generation metric that measures customer service assistance, which converts to sales.
  • First Response Time (FRT) - This key metric, also known as First Reply Time, refers to how long customers wait for an agent to respond.
  • Texts per Conversation - Measures how many SMS texts it takes to resolve a customer issue.
  • Total Number of Chats - Refers to the total number of successful chat sessions from the total number of chats.
  • Unsubscribe Rate - Measuring SMS unsubscribes provides insight into customer satisfaction levels.

Are there unique metrics for work-at-home vs. in-center?

The shift to work-at-home agents resulting from the COVID-19 pandemic, while seismic, doesn't mark a significant change from metrics tracked in-center. The difference lies in the results. Specifically, are remote agents as productive as in-center?

Agent satisfaction is certainly one metric of concern. Work-at-home agents may feel disconnected, leading to dissatisfaction, resulting in a lack of productivity. For the same reasons you measure customer satisfaction, you must measure agent satisfaction and find ways to keep them engaged.

The good news is that productivity among remote workers is generally on par, and in many sectors, better than their in-office counterparts — 47% more according to one study.

Even better, McKinsey says that engaged and satisfied call center agents are:

  • 8.5x more likely to stay than leave within a year.
  • 4x more likely to stay than dissatisfied colleagues.
  • 3.3x more likely to feel extremely empowered to resolve customer issues.

Service Level is another metric to watch. How do you know at-home agents are working and not bingeing on Netflix? Consistency is just as important with remote agents as in-center.

Should we keep measuring AHT, especially in a post-COVID era, with supply chain and customer service issues?

Progressive contact centers that are more "First Contact Resolution" focused no longer rely on AHT. With all the variables that impact AHT—inability to hire and train agents, supply chain issues, more complex problems—it's hard to control both FCR and AHT. Also, with the improvement in self-service, call volume typically consists of a more complex series of calls, which AHT can measure less predictively.

That doesn't mean we throw AHT under the bus, however. As mentioned, it is a valuable forecasting measure. Plus, you can draw a clear line of distinction between a shorter AHT and reduced customer wait times. The downside is that it can incentivize agents to prioritize speed over quality. Indeed, customers want a quick resolution to their issues, but not at the expense of the agent rushing to get them off the line to take another call.

Should we focus on resolving the customer's issue and aiming for high Overall Satisfaction (OSAT)?

Overall satisfaction is a mission-critical metric regardless of the industry. The amount of competition for a company’s goods and services coupled with the customer's ability to rapidly shift to competitors means that satisfaction is far weightier than handling a customer's call within X number of minutes.

As supply chain problems get more complex, customers not only want a resolution, they also want it handled as efficiently as possible. The one thing they do not want is to have to make multiple calls.

One way to gauge customer satisfaction is using surveys that contain OSAT-related questions, such as “What was your overall satisfaction with the experience?” asking them to rate their entire experience on a scale of 1 to 10. That could be accompanied by the NPS question, “How likely are you to recommend [company name] to a friend?” also rated from 1 to 10.

Either way, in an era where competition in just about every industry is tight, creating a best-in-class experience, delighting, not just satisfying, a customer is an imperative.

There is a problem, however, which we can summarize in one word: agents. Call center agents are the brand ambassadors and voice of the customer. They deal with increasingly complex problems, which makes their jobs harder and more stressful. Providing quality training and supportive management eases their burden, leading them to feel more competent and less stressed. Their satisfaction improves, which directly impacts customer satisfaction.

What about new measurements like employee satisfaction scores (internal NPS)?

Two factors bear on this question: First, hiring is difficult with the "Great Resignation" resulting in stiff competition for jobs. Two, labor is one of the most expensive parts of running a business. It's usually more cost-effective to keep your talent, which is why employee satisfaction scores (internal NPS) matter.

Your clients' customers will appreciate a veteran agent’s level of experience and efficiency versus new call center agents using them as practice. Internal satisfaction measurements are a "tell." If employees can't refer friends and family to your call center or BPO, you will eventually see attrition rates rise.

How do BPOs measure client sentiment as a metric?

"Sentiment" is a subjective term that's hard to define. How can you assign metrics to feelings? Satisfaction surveys are one solution.

A lot of BPOs will have their own satisfaction surveys. These may be informal surveys sent after quarterly business reviews or a formal survey sent annually. It's important to know what your clients feel about you, especially if your business is the voice of the customer.

Another informal way to measure client sentiment is the percentage of new business awarded. If they hate you, the client probably won't provide new business aside from organic growth—that’s if they stay with you at all.


In summary, where call center and BPO metrics in a post-COVID era are concerned, some things change, and some things stay the same. Standard metrics like CSAT, FCR, and Service Level will likely always be with us. Even AHT remains useful.

What has changed is the shift from calls centers as a cost center fixated on getting customers off the phone as quickly as possible to a decidedly customer-centric focus where overall satisfaction is the Holy Grail, regardless of the time it takes to resolve the issue.

Yes, customers are more demanding—and supply chain issues and agent staffing shortages don’t help—but with proper training and strong management support, agents can still offer excellent customer experience.

Common Call Center Metrics

The following are standard call center metrics that vendors report to clients monthly.

  • Attrition: The loss of vendor personnel, whether voluntary, involuntary, or due to internal promotions or transfers, and the company does not hire a replacement.
  • Average Abandon Rate: Total calls abandoned (dropped after 5 seconds in the queue) divided by total calls offered.
  • Average Talk Time: The time an agent spends with a caller during a transaction.
  • Average Hold Time: The amount of time agents place customers on hold.
  • Average Wrap-up Time: The time it takes to complete necessary work immediately following an inbound transaction. Tasks often include, entering data, filling out forms and making outbound calls.
  • Average Handle Time (AHT): The sum of Average Talk Time, Average Hold Time, and Average Wrap-up Time for a specified period.
  • Average Speed of Answer (ASA): The total time it takes to answer all the calls, divided by the number of calls, calculated monthly.
  • Customer Satisfaction (CSAT): Refers to the customer's perception of an agent’s service skills and how the agent addressed his or her issue. It is measured weekly by a third-party market research firm. The client's goal is 85 percent or higher “somewhat” or “very” satisfied.
  • Forecast: An estimate of the total number of customer service calls the client expects the vendor to handle on its behalf in the next month. The vendor submits the Forecast at least 30 days before the month it represents.
  • Fixed Forecast Period: The month that the Forecast represents, delivered 30 days before the start of the month.
  • Line Adherence: The number of agents scheduled during a given time of day to be on calls. It measures the variance between the number of agents on calls versus the number of agents planned to be on calls.
  • Service Level: The percentage of calls answered within a specific length of time that the client designates as acceptable.
  • Shrinkage: A workforce management metric that refers to when agents are being paid but are not available to handle customer interactions. There is planned shrinkage, such as taking agents off the floor for staff meetings and training, and unplanned shrinkage, such as an agent calling in sick or on vacation.
  • Show Rate: The percentage of new agents who show up for training on day one.