The contact center is central to digital transformation, but to determine its effectiveness, business leaders must measure several key performance indicators (KPIs). However, one of the problems is that you may not be able to compare your KPIs to other contact center operations.
In its “
Call Center Performance Metrics 2019-20” report, Ozonetel provides insights into the performance of its cloud-based call center service. The report is based on internal service metrics, covering 250 million calls made by 60,000 agents from various businesses. One of the interesting values of cloud-based services for contact centers is that cloud providers can study a large number of calls across multiple locations. As such, they can provide insight into how well a particular contact center is performing compared to others.
The majority of metrics for the contact center are mostly inbound, which includes customer support, problem resolution, and ordering centers. Outbound traffic includes customer contact calls, such as outreach to customers or for marketing.
Inbound KPIs
For 2019, the report shows the average inbound call queue time at about 79 seconds, an increase from 75 seconds in 2018.
The service level is defined as the percentage of calls answered in under 20 seconds, and the financial industry had a 95% service level. E-commerce had a service level of 98%, while the pharmaceutical industry did poorly with a service level of 80%.
A different measurement is how many calls weren’t answered. This measurement produced an average abandonment rate (callers hang up) of about 12% for 2019, the same as in 2018. The abandonment rates measured were as low as 8.4% and as high as 26%.
The average agent handle time, or talk time, was 260 seconds in 2019, up from 188 seconds in 2018. That’s the amount of time required to complete a call. The talk times range varied from 130 seconds to 559 seconds by industry. The talk time can be improved with good management and training, which leads to faster call resolution time and shorter handle times.
The average speed to answer a call was about 3.6 seconds in 2019, a decrease from 6 seconds in 2018. That doesn’t indicate how long a caller has spent in an IVR or call queue. Instead, it reflects the amount of time an agent has taken to provide an answer. The longer the answer time, the more likely that agent training is required or excessive call volumes were encountered but not predicted so too few agents were staffed.
The agent average wrap time is 25 seconds in 2019, a decrease from 29 seconds in 2018. That indicates how long it takes for an agent to complete their data entry and other post-call work. Wrap times that are equal to or below average indicate a well-designed and well-trained call center staff.
Agent Productivity
As far as KPIs for measuring agent performance, the number of calls per day averaged 27, which is normal, for the report period. At an extreme, this KPI hit 40 to 50 calls during busy times.
The average work time for an agent was 7.5 hours, compared to the average 8+ hour workday. The total break time per day was a little over one hour. Each break was about 14 minutes.
Outbound KPIs
Outbound calls are significant, but a lot of people think such calls are robocalls and don’t answer the phone. The report demonstrates that only 46% of outbound calls were answered in 2019, a slight increase from 44% in 2018.
The report indicated a performance difference between manual dialers (45 calls per day) and predictive dialers (115 calls per day) for an average of 90 calls per day. Automated dialers may be more efficient. When the call center is offering high-value or specialized products or services, the success rate for manually dialed outbound calls could be higher since the agent can select the most convenient times to make the calls, thereby increasing the number of calls answered.
Although the KPIs reported are averages, call center management should be aware that the KPIs will vary considerably based on the products or services sold and the industries supported.