Call centers enable customer service teams for many businesses. At their best, they’re efficient, productive, and able to handle thousands of calls per day. Benchmarking your performance with the right statistics, metrics, and KPIs can produce useful insights into the overall health of your call center.
This is the rate at which call center employees leave their jobs over a period of time. According to Nextiva, the average turnover rate is 30-45% per year.
The leading causes of high turnover are stress, repetitive tasks, demanding performance goals, low pay, limited career growth opportunities, toxic work environment, and insufficient training.
You can calculate your turnover rate by dividing the average number of employees at your call center over a period of time by the number of those who left within that same period. Take that final number and multiply it by 100 to get your result.
Turnover may be voluntary or involuntary, so it’s important to account for each as separate metrics.
There are plenty of strategies that call centers can implement to decrease turnover rates, including:
AHT is the average time an agent spends speaking to a customer. It starts from the moment the agent picks up and ends when the inquiry is resolved—not when the customer hangs up. This metric covers actual talk time, the time the customer spends on hold, and “after-call work,” which is the time spent wrapping up a call.
Your AHT should ideally be around six minutes, depending on the sector.
If your AHT is high, your agents either need more training or you may need to update your training resources. Maintaining a low AHT means call agents can speak to more customers and handle more issues.
It’s important to note that while AHT is useful for calculating the amount of time spent on calls, it doesn’t tell you anything about call quality.
CSAT measures how satisfied customers feel with the support they receive. This type of feedback from the customer’s perspective is great for highlighting areas that need improvement.
You can collect this information by presenting a short survey after the interaction, during the call, or by sending a follow-up email after.
According to the American Customer Satisfaction Index (ACSI), the CSAT score benchmark for call centers that operate in the telecommunications and information industry is 63-78%. This is for companies that provide wireless phone services, subscription TV services, video streaming services, and internet service providers.
CSAT scores don’t always tell the whole story, because unhappy customers are more likely to respond to these types of surveys.
However, customer feedback is important and should be taken seriously, especially if you keep receiving the same complaints.
This is a measure of how many calls your interactive voice response system handles.
IVR is designed to accept calls and help callers resolve their issues via self-service options without the involvement of an agent. It also directs callers to the relevant departments.
The strength of your IVR is directly proportional to a strong containment rate of around 60% or higher. If customers can find answers to their problems on their own, there’s less stress on your agents, freeing them up to handle more complex issues.
A good IVR is simple with easy-to-understand instructions for the callers. To achieve this, you’ll need to plan and develop an effective call flow. You should start with broad categories on the main menu that pertain to the most common reasons customers call in.
If your containment rate keeps dropping, it might be a good time to update your system.
Call abandonment rate is the percentage of customers who give up while waiting to speak to an agent. You can’t really track why customers hang up, but it’s often due to the frustration of a long wait time.
To calculate this, divide the number of abandoned calls by the total number of calls handled and multiply by 100.
An acceptable rate is between 5-8%, although what is considered acceptable depends on the sector.
To prevent a high rate, you should consider:
ASA refers to the length of time a caller spends waiting for an agent. It starts when the caller joins the queue and ends when an agent picks up the call.
Your goal should be to connect each customer with an agent within a minute.
Short wait times will have a direct impact on your CSAT scores and call abandonment rate, so the three go hand in hand. It can also encourage customers to keep calling, which results in the overall growth of your business.
On the other hand, long wait times will leave callers feeling discontented and may result in a drop in call volume.
FCR refers to the number of inquiries handled in the first interaction.
Customer satisfaction will naturally decrease with every additional call they have to make to resolve their issue. So, when FCR is high, it saves time for both agents and customers.
You can measure FCR externally or internally. External FCR measurement involves sending a survey to the customer, while the internal measurement involves waiting for up to 30 days to see if they call back with the same issue.
According to the Service Quantity Management (SQM) group, an FCR rate of 70-79% is considered good, while 80% and above is world-class.
As of March 2024, the average hourly pay for a call center agent in the US is $17.75 per hour, according to Indeed.
This amounts to a total yearly salary of about $38,000 for a full-time position.
The range is from $12 on the low end to $25 on the high end, and depends on the company and location. Locations like New York, Chicago, and Atlanta offer the best pay, but also have higher costs of living.
It’s important to note that this figure doesn’t take into account the experience levels of the agents. According to ZipRecruiter, senior call center managers earn upwards of $47 per hour.
Fraud has been on the rise in recent years, and it’s been frustrating to deal with for both agents and legitimate callers.
According to Pindrop, one in every 1,700 call center calls is fraud of one type or another. These harmful calls can also cost businesses as much as $27 million every year.
Fraudulent calls happen for various reasons, from identity theft to attempts to take over an account. Whatever the reason for a fraud attempt, it results in a loss of revenue and damage to the company’s reputation.
Although call agents have adopted security strategies when interacting with callers, fraudsters may still be able to access the victim’s details through social engineering attacks.
Luckily, call centers can use Call Detail Records to detect fraud. CDRs can analyze unusual call patterns and identify anomalies in call duration and frequency. Specific callers can be tracked, and measures can be put in place to curtail any additional fraudulent activities.
This refers to the number of inbound and outbound calls handled by a call center over a specific period.
It covers both calls taken by human agents and calls handled by IVR or self-service options.
The volume of calls handled by a call center depends on the sector and size of the company, so there’s no industry standard or universal benchmark for call volume.
Tracking this metric for your own business will help you discover peak times when you receive the highest volume of calls. This will help determine the number of agents required to maintain operational efficiency, and help you understand the effectiveness of your self-service options.
This metric measures customer loyalty and satisfaction based on how likely a customer is to recommend your product or service to someone else.
Similar to CSAT, call centers can send out an NPS survey after an interaction, asking how likely the caller is to recommend their company to a friend or colleague.
The customer responses from the survey are separated into three groups:
Promoters and detractors are the most important groups, which is why NPS is calculated by subtracting the percentage of detractors from the percentage of promoters.
Any score above 0 is good.
NPS scores vary across industries, and according to a study by Smoke CI, the current global NPS benchmark in general is 43, and for the telecoms industry, it’s 34.
