Metrics are a crucial part of managing any efficient and profitable remote or hybrid call center. Key performance indicators (KPIs) can help leaders gauge and benchmark the center’s performance across many important elements of the customer experience, even when agents are working from home. But the sheer amount of data to be analyzed can be overwhelming. How do you know which metrics to focus on?
To help, we spoke with call center expert Thomas Laird and compiled a list of the 10 most important metrics for any remote or hybrid call centers.
1. Customer sentiment
The overall Customer Sentiment score of a call center is a measure of the average “sentiment” that customers have after calling into the center. Sentiment is either positive, negative, or neutral and a Customer Sentiment score reflects the percentage of callers leaving with a positive sentiment. This allows centers to monitor sentiment to create a benchmark and then work to improve upon it.
Customer Sentiment can also be measured fluidly in real-time through sentiment analytics. This technology is a game-changer because it allows call centers to better gauge and understand the quality aspects of how their agents are performing on the phone.
For example, analytics can track keywords to understand how often customers are calling in with negative sentiment. If 75% of customers are calling in with negative sentiment, that’s an indicator that something has gone wrong within the customer journey. This not only protects a call center from being scapegoated as the cause for poor satisfaction but also helps the organization uncover the problem that’s causing the issues. It could be anything from an IVR or website issue to a major flaw in the product or service being offered to the customer.
Tracking customer sentiment through the life of the call can then go a step further and show how often customers are calling in with negative sentiment but then leave the call with positive sentiment.
Not only does analyzing sentiment help leaders uncover external issues that are impacting customers (like flaws in the product or along the customer journey), but it also helps them get a much more accurate picture of their agents’ ability to provide quality service and improve the customer experience.
While customer satisfaction score and NPS have long been the standard for measuring customer experience, some experts believe that customer sentiment is already beginning to overtake both as the most important metric related to understanding customer experience. Many call center managers continue to track CSAT and NPS alongside sentiment.
Net promoter score (NPS)
To get NPS, customers are asked to rank how likely they are, from one to 10, to recommend a product or service to a friend or colleague. The average score of a company or service provider is known as their Net Promoter Score. Anything less than nine or 10 is considered a failing score.
Customer satisfaction (CSAT)
Customer satisfaction, or CSAT, is another survey-based measure of how satisfied customers are that focuses on specific aspects of the customer journey. This allows a call center to ask specific questions to better understand how and where the customer experience can be improved.
2. First call resolution (FCR)
First call resolution tracks the percentage of calls that are “handled” the first time that a customer calls in about a certain problem. This is an important and telling metric because it effectively measures both the efficiency and the quality of how calls are being handled. A low FCR percentage means that customers are having to call in multiple times in order to have their issues resolved. This increases call volumes and reduces customer satisfaction.
Using new analytics technology, call centers can tell when a customer is calling back in about an issue more quickly than ever before. In the past, centers relied on lots of manual searching to understand how often callers were reaching out about any given issue or instance of an issue. Using this data to gauge FCR helps leaders understand why customer issues aren’t being resolved on the first attempt and train their staff to be better equipped to do so in the future.
3. Service level
Service level refers to the percentage of calls that are answered within a specific length of time that has been designated as acceptable. For example, a service level of 80/30 means that 80% of calls are answered within 30 seconds or less. This stat allows call center leaders to use analytics to decide how many agents need to be working in order to achieve a certain service level.
While a service level goal of 80/30 is fairly standard across call centers, individual organizations can adjust service level requirements depending on the industry that they’re in and the expectations and requirements of their customers. For example, if a particular customer base is willing to spend more time in the queue before their call is answered without any negative impact on their sentiment, that call center can plan for a longer wait time or a lower percentage answered within that time frame. This allows them to adjust staffing levels to maximize occupancy and, therefore, profitability.
4. Average speed of answer (ASA)
Average speed of answer is the average amount of time that it takes for a customer’s call to be answered by an agent. The purpose of this metric is to show how long customers are typically waiting on hold or in a queue before being helped by a customer service representative. Since long wait times are a known cause of customer dissatisfaction, average speed of answer is a major indicator of how customers are likely to feel about their experience with the call center. Forecasting accuracy, adherence to schedule, and occupancy play a significant role in being able to manage ASA effectively.
5. After call work / wrap time
After call work, also known as wrap time, refers to the length of time that agents spend completing tasks associated with a particular call after hanging up the phone. This measurement typically begins as soon as the call ends and continues until the rep has finished any memoing or form completion required. There is usually a specific amount of time allotted for after call work so the purpose of measuring is to make sure that agents are staying within that allotted time. A good benchmark for ACW is 30 seconds.
If agents are consistently exceeding their allotted time for after call work, then average speed of answer and the number of calls that they can handle decreases, and service levels drop. After call work is considered by some experts to be the #1 killer of efficiency since it has such a dramatic knock-on effect on many other efficiency metrics.
For centers experiencing long ACW times, there are some tricks that they can use to improve them. Start by streamlining processes, incorporating shorthands for memoing like using “CU” instead of the word “customer,” or training agents to be more efficient notetakers.
6. Average handle time (AHT)
Average handle time is the average length of time that it takes a rep to complete a call. This measurement begins the moment an agent answers a call and is counted until they complete all after call work. Average handle time can vary across industries and customer bases so it’s important for call centers to find their ideal AHT and benchmark agents off of that.
When average handle time starts to increase and exceed benchmarks, it can be a good indicator that something has gone wrong in the customer journey. This could be something like issues with the IVR or even with the product itself.
Another possible cause for long AHT is inefficiencies in the way calls are being handled. For this, centers can help by doing things like implementing technology to more quickly verify customer information or evaluating new ways of reading disclosures.
7. Abandoned time
The percentage of abandoned time is the percentage of calls that are abandoned before they can be answered. The idea behind this metric is to help call centers better understand how often callers are hanging up due to long hold times. This is a great way to quantify when average speed of answer is too slow. When customers are abandoning calls because they are not being helped quickly enough, it’s likely that they will be dissatisfied with their experience. A common benchmark for abandoned time is 3% of calls or less.
There is some nuance to this metric as some calls will be abandoned for other reasons than long wait times. Whether it’s because they made a mistake in the interactive voice response (IVR) system or simply decided they no longer wanted to call, some customers will hang up before an agent has a real opportunity to answer them. It’s important to take this into account when calculating abandoned time. Some call centers won’t include calls that hung up within 5 seconds or less in their abandoned numbers.
8. Forecasting accuracy
Forecasting accuracy is a metric that shows how well a call center is forecasting when their staff is needed. Call center supervisors and workforce management (WFM) teams use historical data and call volume trends to determine how many agents are needed in order to maintain acceptable service levels. This is sometimes done up to two or more months in advance to ensure a fully staffed team to handle expected call volumes. Forecasting accuracy measures how accurate those forecasts are.
Forecasts, as well as forecasting accuracy, are constantly changing. An ideal benchmark for forecasting accuracy is for your actual need for agents to be within eight to nine percent of your forecasted staff numbers. The Erlang C. Calculator is a helpful tool that call centers can use to determine how many reps per interval they will need based on call volume, handle time, and desired service level. It can help them improve forecasting accuracy, or it can be used to determine how they can adjust service levels to better fit their staff level capacity.
9. Adherence to schedule
Adherence to schedule is a measure of an agent’s, team’s, or entire workforce’s percentage of actual time worked compared to expected time worked. Expected time worked is the amount of time, minus lunch and other scheduled breaks, that an agent should work in a given day and adherence to schedule is the percentage of that time that an agent actually works. The goal is 100%, however, a certain percentage of “shrinkage” is assumed. This is lost time due to typical and acceptable interruptions in a workday.
Disruption in adherence to schedule can cause major service issues, so it’s an important metric for workforce management (WFM) to watch. They typically use a report that will show agents as “out of adherence” if they aren’t meeting expectations. When adherence to schedule is lacking, WFM will notify supervisors to remedy the situation with the rep.
10. Occupancy / utilization
Occupancy is defined as the percentage of time per hour that a rep is handling phone calls. This is different from adherence to schedule in that it’s specifically measuring an agent’s handle time (including after call work), while also factoring in average speed of answer. Where adherence to schedule measures actual work compared to scheduled work in a given day, occupancy measures actual minutes spent handling calls per hour. A standard benchmark for occupancy is around 70 to 75% per hour or around 45 minutes per hour. This is because very high occupancy rates near 98 and 99% can lead to agent burnout.
This metric is useful for call centers to maximize profitability through more efficient scheduling. For example, if occupancy is only at 50%, then the center is considered to be overstaffed and inefficient. There is an opportunity to be more profitable by reducing the workforce so that the agents on staff are more occupied. For outsourced call centers being paid by the minute, this is especially important.
There’s a lot to measure when it comes to call center metrics. Everything from how long it takes to answer a call to how long it takes to finish memoing when the call is done. While all call center metrics can provide insights into how a call center is performing, these 10 are considered the most important for any remote or hybrid call center.
For more detail and insights from call center expert Thomas Laird, check out our Ultimate Guide to Call Center Metrics here.
Ryan Plank is a content marketer with a degree in Journalism and a background in technology. He lives in Orlando, Florida, and is an avid golfer.