What are contact center metrics?
Contact center metrics are indicative performance data that quantify the efficiency, experience, and satisfaction of your customer service. In short, they track the overall quality of your support, but the quality is only the tipping point. Contact center metrics and key performance indicators (KPIs) also estimate your contact center’s customer satisfaction cost and revenue. As any contact center would have it, part of the manager’s role is to show high-quality support with a lean team and cost-efficient financial KPIs.
Top contact center metrics to track
Contact center metrics can be grouped into five broad types based on the actionable business and customer insights that they provide. From the different types of contact center metrics, let’s drill each of them down to what insights they stand for and how they affect the customer experience and organizational goals.
Customer service metrics
Customer service metrics represent the quality metrics of a contact center. They are directly tied to the performance of your support agents and the explicit or implicit feedback that customers indicate about their support experience.
1. Customer satisfaction (CSAT)
Customer satisfaction or CSAT score is one of the primary metrics used to determine customer service quality levels. It is directly collected from your customers after interactions with your support team, and is the easiest way to evaluate and improve your customer experience. Different types of surveys from simple rating-based forms to interactive voice response (IVR) surveys are used to collect CSAT responses from customers.
How to calculate your CSAT
Here, ‘no. of positive responses’ corresponds to the number of customers that provided a rating of 4 and above out of 5.
Consider that out of a total of 100 survey responses, 62 of them have provided a rating of 4 or higher, out of 5. Your CSAT score would then be (62/100)*100 = 62%.
Is your CSAT score satisfactory?
CSAT scores anywhere between 75-85% can be considered as a great standard, even though this might depend on a variety of factors including the industry and the size of your business. One more thing to consider is that this scoring method only includes your promoters (people that have rated 4 and above, that are active brand advocates), so it might be impossible to achieve a near-perfect score.
Average CSAT Score: Benchmarks by industry
To see some real-life examples of customer satisfaction scores, let’s look at the average CSAT scores across different industries.
Automobiles and Light Vehicles: 78
Cell Phones: 79
Computer Software: 76
Consumer Shipping: 76
Credit Unions: 77
Financial Advisors: 77
Full-Service Restaurants: 79
Internet Retail: 78
Internet Travel Services: 74
Life Insurance: 78
Specialty Retail Stores: 77
2. Average waiting time
Waiting time is defined as the duration spent by customers waiting in a queue before they get to talk to an agent. If they had to move between multiple queues before actually talking to an agent, the sum total of all the waiting times is considered for calculation.
Keeping your waiting times low is critical in customer service since it has a huge influence on satisfaction and overall customer experience. It also serves as a direct indicator of how efficient your operations are.
How to calculate average waiting time
Say you have received 400 calls in total over a week, and the time spent by customers waiting on the call for your agents sums up to 1200 minutes. Therefore, your average waiting time would be 1200/400 = 3 minutes.
Is your average waiting time satisfactory?
Most experts agree on the 80/20 rule for call waiting, which is 80% of the calls need to be connected within 20 seconds for a highly satisfactory customer experience. However, the latest research reveals that this is not a standard measure and can vary by industry and even based on the brand identity.
As a rule of thumb, you should calculate other metrics like call abandonment times (the duration for which callers wait before they disconnect) for your contact center, and then, aim for a corresponding waiting time.
For example, if 30% of your callers abandon the call till 60 seconds after calling, and if there are other measures (such as callbacks, callback requests, deflection etc.) to handle this 30% dropoff, then it would make sense to aim for an average waiting time that is less than 60 seconds.
3. Call abandonment rate
Call abandonment rate is the ratio of calls abandoned by customers and inbound calls completed. Call abandonment happens when agents put customers through cold transfers and long hold times while addressing an issue. Customers choose to drop off the call since they lose patience and belief in the resolution process.
How to calculate your call abandonment rate
Before working out your call abandonment rate, there are a few metrics that you need to understand in order to calculate it accurately:
Active calls: total number of ongoing calls at any given time
On-hold calls: number of calls/callers that are in a queue waiting to speak to an agent (can be after a conversation with IVR, or even another agent)
Abandoned calls: the number of callers who hang up when waiting in a queue without speaking to a live agent, after interacting with the IVR system or with a forwarding agent.
Consider that your contact center receives about 1000 calls, from which 50 are abandoned in total. Therefore, your call abandonment rate would be (50/1000)*100 = 5%
A call abandonment rate of 12% is widely accepted as a standard across industries. However, with an increasing number of mobile callers to customer support, rates as high as 20% are being accepted by contact centers as a benchmark.
4. Net promoter score (NPS)
NPS is a game of customer stickiness, loyalty, and advocacy. Do you think your customers would recommend your product and the overall support experience to their peers? Crack that code with proactive surveys and you have an answer.
How to calculate your NPS
In this case, promoters are users that give a rating of 9 or more out of 10 (active, loyal customers of your brand), and detractors are the ones that give any rating equal to or less than 6 (unhappy customers that can potentially spread negative word of mouth and damage brand reputation).
Consider 100 survey responses, split into 70 promoters, 20 detractors, and 10 passives. Here, the NPS score would be (% of promoters) – (% of detractors), which equates to a score of 50 (70%-20%).
5. Customer effort score (CES)
Customer effort score or CES is the amount of time or ease with which a customer is able to find appropriate support information and get issues resolved. The metrics to measure CES can vary for different companies but ironing out the customer journey is the difference between low effort and high effort.
How to calculate customer effort score
There are multiple methods, such as the 1-10 scale, Likert scale, and the Emoji scale, to collect feedback from users, and the formula to calculate CES from them varies based on the method used.
1-10 scale: the user is posed with a direct question such as “How easy was it to get your issue(s) resolved?” with a response range from 1 to 10, and any responses above 7 are considered as good standard.
Likert scale: this system has a standard 5- or 7-response template, where users can answer anything between “Strongly agree” and “Strongly disagree”.
Emoji scale: this method mostly uses three emojis — happy, neutral, and sad — to gauge customer satisfaction, and the average score is calculated based on the percentage of happy and sad emoji faces recorded.
“How happy were you with this resolution? Please select a response.”
Please provide an appropriate response.
I found it easy to solve my issue through your customer service portal.
How happy were you with this resolution?
6. Preferred channels
Customers’ preferred channels can be determined by analyzing the ticket volume across channels for different types of issues. For instance, a refund query might find its way to you through an email while a network downtime triggers social mentions and complaint tweets. Understanding customers’ preferred channels of support can help customer service managers forecast workload and distribute agents accordingly.
7. First contact resolution (FCR)
First contact resolution is the ability to close out a customer query within the initial response or contact. A comprehensive knowledge base and a 360-degree customer view can help agents diagnose and resolve issues instantly on first contact. For call centers, FCR is more focused on first-call resolutions, whereas, with contact centres that handle customer support on multiple channels, first-contact resolution is accepted as a standard metric.
How to calculate first contact resolution rate
Consider that over a period of 90 days, your contact center has received 600 tickets, and 120 of those were resolved within the first interaction with your support team. The FCR rate then works out to (120/600)*100 = 20%, which means, 20% of all your incoming tickets were resolved during the first interaction itself.
8. Escalation ratio
Escalations are customers’ way of signalling that they are unhappy with the support provided. This could either be a result of delayed support or insufficient resolution. Customer service managers need to follow an escalation matrix for a clear flow of actions to pacify the customer with a favorable solution to their problem.
How to calculate escalation ratio
Similar to the example mentioned above, if there are 600 tickets over a period of one month where 60 tickets were escalated, the escalation ratio for the month can be calculated by (60/600)*100 = 10%.
9. Customer issues by type
Identifying the top contact drivers to your customer service can signal improvements in forecasting and bring readiness to your support agents. To measure this, you need to tag every customer query by ticket type and create reports based on the most common types that are recurring.
Examples of ticket tags in the retail industry:
Refunds and returns
10. Average call transfer rate
Call transfer rate is the average number of calls that get transferred from one agent to another during a support request. While call transfers might make sense for certain complex situations, they can end up a dealbreaking experience for customers. Customer don’t want to repeat themselves all over to another agent, or wait on cold transfers. Skill-based routing is an intelligent way around this common issue, where requests are assigned to agents based on their expertise.
How to calculate average call transfer rate
Here, ‘first live touch’ refers to the first interaction the customer has with a live agent (excluding forwarding agents).
If a call center handles about 1500 calls in a month, and 150 of those calls were required to be redirected to another agent/department after first contact, the average call transfer rate here can be worked out to (150/1500)*100 = 10%.
Productivity metrics, unlike quality metrics, are more inclined towards the internal processes of a contact center. They represent the support efficiency and customer handling speed of a customer service team. Agent productivity metrics depend on the leadership, operations, and more importantly, the contact center software or tech stack deployed.
1. Resolution rate
Resolution rate is the number of tickets successfully closed over a defined timeline. SLAs and KPI timelines vary between industries and organizations of different sizes. Based on organization’s ideal reporting timeline, you can measure the resolution rate per day, week, or month.
How to calculate resolution rate
If a call center resolved about 800 tickets in the past month from a total of 1000 received tickets, the resolution rate is calculated as (800/1000)*100 = 80%.
2. Average handle time (AHT)
Average handle time or AHT shows the amount of time your agents spend on a support ticket, across channels. In call center speak, call handle time (CHT) is a good indicator of support efficiency as phone is usually a high priority channel where customers bring their complex and urgent issues.
How to calculate average handle time
Let’s assume a contact center spent a total of 8640 minutes on 720 tickets and calls throughout a month, inclusive of actual call and conversation durations, hold times and post-call resolution time. The average handle time can be calculated as 8640/720 = 12 minutes per ticket.
3. Average first response time (FRT)
Average first response time or FRT is the time taken to reply to a customer query. The response may or may not be the resolution itself, but an acknowledgement of the customer’s issue. Conversational bots and IVR are a great way to automate first-response and collect customer context even before an agent picks up a request.
Automated messages can be set up outside of working hours in order to reduce this first response time. Here’s a quick example of what these messages would look like:
“We’re away right now but our agents will be back by 9am tomorrow. Meanwhile, would you like to elaborate your request so that we can work on this first thing in the morning?”
How to calculate average first response time
If we consider total wait times (before first interaction/response) across 150 tickets in a contact center to be 75 minutes, the average first response time for that period works out to be 75/150 = 0.5 = 30 seconds.
4. Average resolution time
Customers want speedy resolutions with less time spent in establishing the problem and context. Resolution time is the metric contact centers need to monitor to assess agent quality, documentaion, and operational efficiency.
How to calculate average resolution time
With a total time of 900 minutes spent across 150 resolved tickets, the average resolution time can be calculated as 900/150 = 6 minutes per ticket.
5. % SLA adherence
Service level agreements (SLA) are set in stone to meet customer expectations. It’s important for contact centers to abide by the promised support timelines to earn trust. The SLA can vary between industries, company size, and also channels. But the key is to define them and make it clear for the customers without any ambiguity.
Here’s how an SLA workflow would look like:
How to calculate SLA adherence
Consider that 400 tickets were received in total, out of which 320 were resolved within the corresponding SLA criteria. This can be worked out as a SLA adherence percentage of (320/400)*100 = 80%.
6. Average number of zero-touch resolutions
Zero-touch resolution is a relatively new contact center metric that will eventually define effortless customer service. Zero-touch resolutions are the ones that are automation and AI-driven. This can be achieved through self-serve bots, peer communities, and a comprehensive knowledge base.
How to calculate average number of zero-touch resolutions
If the total number of resolved tickets amounts to 300, out of which 150 were resolved with zero touch, the average number of zero-touch resolutions comes to 150/300 = ½ which means, one out of every two cases is being resolved with zero touch.
Additionally, consistently measuring the traffic and engagement on solution articles and communities will help earmark the impact of self-serve portals in zero-touch resolutions.
7. Average lifespan of open requests
Determining the average lifespan of open or unresolved requests gives a peek into the real-time productivity and efficiency of your support team. By knowing the timeline or duration through which a ticket remains open, you can go beyond historical data and fine tune the ongoing support strategy.
How to calculate average lifespan of open requests
With a total duration of open tickets amounting to 2400 minutes for 100 requests, the average lifespan is calculated to be 2400/100 = 24 minutes.
Cost and revenue metrics
Contact center is fast being recognized as a revenue center over the traditional perception of a contact center. Cost and revenue metrics help contact center managers make a strong case to the leadership about the bottomline performance of their customer service team.
1. Cost per call (CPC)
Cost per call is the operating cost of a customer call in a contact center. This is an important metric that determines the cost-efficiency, resource allocation, workload, and the agent performance in a contact center.
How to calculate cost per call
Consider a call center incurring $100,000 a year as its running cost, with a capacity of handling upto 80,000 calls.
The cost per call would then be 100,000/80,000 = $1.25 per call.
2. Ticket to order ratio
Ticket to order ratio is usually a north metric for direct-to-customer (D2C) companies like retail and ecommerce. But it can be adapted to other industries too, with a 2:1 rule. It’s a good practice to maintain a flow of no more than 2 tickets per every customer order, to keep your contact center lean and mean.
How to calculate ticket to order ratio
Say you receive about 300 orders in a period of one month, and the total number of tickets in this period is 450. So, the ticket-to-order ratio here is 450/300 = 3/2, which means there are 3 tickets created for every 2 orders placed with your company.
3. Customer retention rate
Customer retention is the percentage of customers who remain with your business over a defined period. This is a contact center metric that hugely impacts boardroom conversations and decisions, and gives an opportunity to examine what’s working and not working in the customer service department.
How to calculate customer retention rate
Consider that a business had 3000 customers, and at the end of their previous quarter, got 300 new customers. At the end of the quarter, they had a total of 3200 customers. In this case, the customer retention rate is calculated as [(3200-300)/3000]*100, which is approximately equal to a 97% retention rate.
4. Customer churn rate
Customer churn rate as a metric serves a similar function as customer retention. It’s the percentage of customers who stop buying or subscribing to your business over a defined period.
How to calculate customer churn rate
Consider that a business initially had 3000 customers, and at the end of their previous quarter, had lost about 300 customers. Then the customer churn rate would be equal to (300/3000)*100, which results in a churn rate of 10%.
Agent experience metrics
Agent experience metrics are an extension of employee satisfaction, which is vital to an engaged, customer-first contact center. Agent experience metrics help contact centers look inwards and fine-tune processes that affect employee happiness. Workload, for instance, is a major factor that drives agent experience metrics.
1. Agent attrition rate
Agent attrition is the rate at which contact center agents leave their job, over a defined period. High attrition can severely impact the normal functioning of a contact center with increasing hiring costs, long training windows, workload management, and more importantly, disillusioned customers. It makes a strong case for every companies to invest in agent empowerment.
How to calculate agent attrition rate
If a customer service team started with 300 agents and if 15 agents left during the course of a year, the agent attrition rate for that year would be (15/300)*100 = 5%.
2. Agent absenteeism
Agent absenteeism, otherwise known as absence rate, is an indication of the work culture of contact centers. Workload and stress are directly related to unexplained and unplanned absences at work. It can be result of micromanagement, extreme workload, poor employee culture, or insufficient agent enablement strategies.
How to calculate agent absenteeism
Considering that a customer support team has 2500 working hours planned for all of its members included, a total absence of 100 hours would amount to an agent absenteeism rate of (100/2500)*100 = 4%.
3. Employee NPS
Similar to customer NPS, employee NPS is used to take a pulse of what your agents feel about your organization and contact center teams. The best way to get employee NPS is by sending surveys and collecting their anonymized feedback for various questions.
When we think contact centers, we usually imagine a call, email, or message sent from a customer to a company. But some customer service teams also double-up as an outbound contact center to proactively engage for support and sales. These metrics denote the performance of business or agent-initiated customer conversations.
1. Rate of successful callbacks
Not every customer call goes through to an agent instantly. Many contacts drop off without waiting, some try IVR to get their issues resolved, and a few opt for a callback. Rate of successful callbacks is the average number of times an outbound call is made or picked up by the customer.
How to calculate rate of successful callbacks
Consider 80 successful callbacks with a total callback request count of 100, the rate of successful callbacks can then be calculated as (80/100)*100 = 80%.
2. Percentage of support-driven upselling and cross-selling
Sales isn’t exclusive to cold outreach emails and calls from sales folks and account managers. A huge chunk of a company’s sales can come from the subtle, yet valuable problem-solving agents indulge in daily. Customers also have better relationships and trust with support leaders who don’t necessarily sell but offer tangible solutions.
How to calculate percentage of support-driven upselling and cross-selling
In a total ticket count of 100, if 25 tickets resulted in some kind of upsell/cross-sell resulting in increased customer value, the percentage of support-driven upselling and cross-selling would be (25/100)*100 = 25%.
3. Number of proactive support calls per month
Agents make proactive support calls to let customers know about issues before they actually arise. This could be about an account problem, network outage, feature requests, or even post-purchase advice.
4. Call attempts per customer
Call attempts per customer is the number of times an agent or a team try to contact a customer for sales. It is a good indicator of the customers’ interest or the lack of it in lapping up cross-selling and upselling requests from agents.
Real-time contact center metrics are the future
Handling customer requests without real-time contact center metrics and performance monitoring is like driving a sports vehicle without a dashboard. The functionality to make support decisions on the go and course correct can be a competitive advantage that separates a modern, customer-aware contact center from a conventional one.
Benefits of real-time contact center metrics
Consistent customer service from readily available historical data and AI-powered agent assistance
Identifying agent strengths and weaknesses with real-time insights into performance
Flexibility to adapt to ticket trends, fetch insights on their causes, and reduce future tickets
Live compliance monitoring to ensure agents are using brand-approved guidelines and communication
Complete visibility on SLA timelines and escalation-prone cases with AI-led early warnings
Achieve industry-leading contact center performance with Sprinklr
Sprinklr’s contact center software helps you monitor and improve your metrics without putting customer context on the back burner. Built for agents who love providing world-class support to their customers, Sprinklr provides context-rich live performance dashboards that steer support teams to spring to action.
Keep track of ticket trends, CSAT, call metrics, SLAs, and customize your agent desktop view to prioritize the metrics that matter the most to your business. Built on top of the world’s only unified customer experience management (Unified-CXM) platform, Sprinklr Service
Improve agent performance and compliance by monitoring every conversation with Sprinklr AI
Drive business-critical intelligence in real-time with Sprinklr’s always-active AI engine
Inform better agent assistance and conversational AI with actionable insights on support gaps and customer pain
Unify data from 30+ social and digital channels to feed your contact center metrics
Leverage contact center automation to reduce agent workload and provide more first-contact and zero-touch resolutions with IVR, chatbots, and knowledge base.
What are contact center metrics?
Contact center metrics are the performance data associated with the functioning of a customer service team. The metrics showcase insights from the results of agent-customer conversations to drive better support and business decisions.
What is the difference between call center metrics and contact center metrics?
Call center metrics are usually restricted to a brand's voice or phone support. Contact center metrics are more holistic and provide insights from modern channels such as social media, instant messaging, email, live chat, and voice.
What metrics do we need to track in a contact center?
The primary contact center metrics can vary between different industries. But if we are talking about the most quintessential metrics, here are some: customer satisfaction (CSAT), call abandonment rate, net promoter score (NPS), average wait time, first response time (FRT).
What can you learn from contact center metrics
Beyond the primary use of monitoring agent performance and customer satisfaction, contact center metrics can help fetch product insights, reduce contact volume, common customer issues, operational gaps in customer, and hidden customer cues to name a few.
What are the types of contact center metrics?
Based on the agent, customer, and business insights that they inform, contact center metrics can be classified into:
Customer survive metrics
Cost and revenue metrics
Agent experience metrics
Do you provide a free trial of Sprinklr Service?
For Sprinklr Service, we provide a thoroughly personalized demo since we believe that enterprises have unique needs and a customized approach to getting started. But if you are in for Sprinklr Service, we provide a fully functional 90-day trial with no fine print or credit card frills. Experience Sprinklr’s ground-breaking contact center software which is purpose-built to make customer service simple, contextual, and efficient.
Thank you for contacting us.
A Sprinklr representative will be in touch with you shortly.
It's time to discover how a unified strategy can help you make customers happier.
Contact us today, and we'll create a customized proposal that addresses your unique business needs.
Request a Demo
No need to fill out any forms — you're all set.