Customer service may rely heavily on the human element, but that alone isn’t enough. Companies need to identify and track the right customer support metrics.
You see, customers have expectations: personal, professional, fast, and efficient service. How do you know you’re meeting these expectations if you’re not tracking the right metrics to get the data? The short answer: you don’t; you’re only left guessing.
Support metrics help you understand how your customer service team is performing, how satisfied your customers are, and where you need to improve.
But, deciding what support metrics to track can be tricky. To help you along, here’s a curated list of the nine top metrics and the best practices to follow when tracking them.
Nine top customer support metrics to track
Resolution time is the average amount of time it takes a support team to solve a customer problem, from when a ticket is opened to when it’s marked as closed.
Longer resolution times generally lead to unhappier customers, so you need to keep your resolution times down without compromising response quality.
A high-resolution time could suggest that your team needs more training to deliver customer service or use specific tools. They may also have too many responsibilities making it difficult to resolve all tickets on time.
Your system for logging and closing tickets may also need work. Perhaps there’s confusion with multiple people replying to tickets, slowing down the entire process. Or, maybe support tickets are not being filtered through to the right people.
To quickly filter messages to the right people, consider using software that lets you easily assign messages to support agents with the right expertise and knowledge to handle them.
Resolution rates
The resolution rate compares the number of requests assigned to a company or representative to those resolved for a certain period.
You can calculate resolution rates with this formula:
Solved Requests / Received Requests x 100 = Resolution Rate
For example, if an agent were assigned 90 requests in one month and solved 80, their resolution rate would be roughly 89% (80/90*100).
Knowing your resolution rates helps determine if you’re meeting your service level agreements (SLAs) and if your customers are getting the service they deserve.
An increase in resolution rates suggests you should keep doing what you’re doing. Don’t fix what ain’t broke, right?
A decrease could cause concern, and you’ll have to investigate why. It could be due to poor customer support workflows, causing requests to be assigned to the wrong people, or being understaffed, causing some to slip through the cracks.
In these cases, your team may benefit from a shared inbox. A shared inbox provides a shared view of email, allowing teammates to assign messages to the right people, collaborate on support requests for faster responses, and prevent multiple or no replies because everyone can see who has responded.
You can also adjust shifts to distribute the load effectively by moving team members around to handle requests during off- and on-peak periods.
Average first response time
First response or first reply time shows how quickly an agent responds to a customer service request when the customer logs a complaint. Autoresponders do not count towards this metric.
But, how fast should a response be? What is acceptable?
As a general rule, faster response times are better—even if it’s a short email. Just avoid the standard email responses and, instead, "Wow," customers a little with some personalization and personality. These replies show customers you care and that their problem is being addressed.
The speed of the response will vary depending on the communication channel. For email requests, a response time of 24 hours is usually acceptable. For social media support, anything less than 2 hours is fine.
That said, speedy responses shouldn’t come at the expense of response quality. If your team is constantly under pressure to respond immediately, the quality of your responses may suffer.
So be sure to set a balanced response target your team can meet. Finally, communicate that target to your customers so that you effectively manage their expectations from the first point of contact.
Related reading: 3 customer support metrics to watch to cut response time.
Reaction time
Reaction time is not the same as response time. Customer service representatives may need to investigate an issue further before responding to the customer.
This metric measures how fast a customer rep actions a message, whether tagging, reassigning, or responding to it.
This task may seem easy enough, but when reps receive a lot of tickets, aren’t adequately trained, or don’t have proper systems to action messages, things can become challenging—fast.
The result? Emails invariably fall through the cracks.
An easy way to address this problem is to change your team’s inbox sorting system so that messages are prioritized, grouped, and assigned based on specific tags.
For instance, your software support team could prioritize messages based on high or low urgency according to specific keywords appearing in the subject line or body of the email:
Tag with high priority if the email contains the word "glitch."
Tag with low priority if the email contains the word "feature request."
Ticket volume
Ticket volume is the total number of tickets received during a specific period. It provides valuable insights into your service quality or other underlying issues.
For instance, high ticket volume around a particular topic like a specific product may indicate a product issue that needs fixing or help you identify repeated questions that get asked.
Instead of your support team answering all these questions, you could create an FAQ page or self-service knowledge base for customers to help themselves. These resources will lighten the load for your support team to focus on delivering better service to other customers.
Ticket volume can also help you pinpoint busier (or slower) customer service times. Use this information to plan shifts and distribute the workload across your team to meet expected demand.
Conversations handled per rep
You need to benchmark how many support requests you’ll assign to different reps based on their experience. More experienced agents will handle a larger volume of requests compared to juniors.
Having these benchmarks and tracking them so you’re sticking to them ensures you don’t overload team members. It also helps guide your internal workflows and systems for assigning new messages as new customer service employees are onboarded into your support team.
Replies per ticket
Replies per ticket measure how many responses, on average, it takes to resolve a ticket. A lower number is preferred as it suggests your team is more efficient in resolving problems, leading to happier customers.
A higher number tells you there was lots of back and forth interaction before closing the ticket. There may be some underlying problems to look out for.
If replies per ticket for a specific agent are constantly high over time, it may indicate the agent needs more training and tools to do their job properly.
If you notice a one-off-isolated event, it’s likely due to other factors. Maybe the problem being addressed is more complex than usual. More input may be required from internal teams and the customer to get to the bottom of it.
If the replies to resolve are high for your entire team, it may even suggest a problem with your customer support system. Perhaps there’s no clear ownership over emails because of an inability to assign them to an owner, leading to multiple replies from the team.
It could even be that messages are passed on without any background context, leading to agents having to ask the customer for clarification without a support agent to clarify.
Again, a shared inbox can help, allowing you or agents to assign clear ownership to an email so everyone knows who must respond. A complete history of background information is readily available, so reps don’t have to do much behind-the-scenes like digging or frustrate the customer with more questions.
CSAT (Customer Satisfaction)
CSAT measures how satisfied customers are with your product or services. Higher satisfaction generally contributes to higher customer retention and profits.
Companies usually send surveys to customers immediately after an interaction with an agent to rate the customer’s level of satisfaction. The survey includes multiple questions, with respondents recording their answers on a 5-point scale (5 is very good and one is poor).
Actual questions will vary between companies, but examples may include:
How would you rate the service just received from the agent?
How helpful was the agent in fulfilling your request?
These surveys provide you with quantitative data to assess current satisfaction and take any corrective action if needed.
To get your customer satisfaction score, record all responses of 4 or 5 and divide them by the total responses. Then, multiply the ratio by 100 to get the percentage of satisfied customers.
Total Breaches
Total breaches is a help desk metric that measures the total number of support tickets that did not meet specific SLAs. For example, if it’s company policy to respond to a customer request within 24 hours and you only respond to 50 out of 100 during that time, you’re in breach of your SLAs 50% of the time.
Keep your numbers down by having well-trained agents, proper workflows, and the right software. For example, Front’s shared inbox lets you tag messages approaching breach to ensure they get prioritized accordingly.
Best practices for tracking customer support metrics
Create reports for easy support metric tracking
Choosing what metrics to track is one thing. But, how do you effectively track them? One way is through reports that allow you to view all your important data at a glance to make those swift and vital decisions.
For example, Front lets you measure email volume, CSAT score, resolution time, response time, and other email trends for each agent or across your team in customizable analytics reports. Use the data to decide when to hire, what training is needed, and staff shifts.
Deliver the reports to those who need it
Your support team does not operate in a vacuum. The data your team gathers from tracking these metrics should be transferred to department leaders in sales, product, success, and other relevant teams so they can act on it.
For example, if there’s suddenly an influx of tickets around a specific product, the product development team will want to know. It will provide insights into product improvements and fixes. Similarly, common client questions are crucial data the sales team can use for their sales materials and prepare for presentations or client objections.
Inspire and challenge the team with healthy competition
The success of tracking metrics will depend on your team’s involvement and input. Get their buy-in by explaining the "why" behind tracking metrics and how it can help them become better agents. Tying excellent service to a performance bonus is always a good incentive.
Finally, you can also incentivize and challenge everyone to participate by creating a competition. For example, you could pick one specific metric like resolution rates and mention that agents with the best resolution rates each month get a prize of $500.
Use customer support metrics to deliver better service today
The human component will always be part of providing amazing customer service. But hard data will give you a leg up on the competition.
It’s the secret sauce that tells you what you’re doing well and what you need to improve. It tells you whether you’re slow to respond to customers, how long you take to resolve requests, and how often you’re not meeting SLAs.
Armed with that data, you can identify and implement solutions to improve service. Maybe you’ll need to improve support workflows or train employees. Perhaps you’ll need to be smart about how you schedule shifts or invest in better software and tools like shared inboxes.
Regardless of what you do, remember to challenge your team. Inspire them with healthy competition, share data with other departments who need it, and invest in the right tools to track metrics and create reports.
Written by Nick Darlington