If you’re the manager of a support team, you have a billion things to think about every day. You need to make sure your team feels supported, then deliver metrics and information to the executive team, and, every now and then, answer a few escalated tickets yourself. It’s a lot.
How, with all of that, are you supposed to find out what metrics are the right ones to report? And, how can you know if they’ll help you grow?
There are metrics that influence the entire company. However, to get insight into your customer service, focus on those that have an impact at the team-level, and allow you to grow your support offering as quickly as possible.
Here is a list of the top six customer service metrics to focus on. Learn how to scale them as your team and company start to get bigger and your needs shift.
Prefer watching videos instead? Scroll down, we’ve got your covered.
Customer Satisfaction Score (CSAT)
CSAT is short for Customer Satisfaction. CSAT is used to measure how satisfied your customers are with your support, or the service or product that your company provides.
According to Qualtrics, CSAT is usually measured by using variations of the question:
“How would you rate the support you received?”
This question usually comes with a scale. Respondents are asked to rate the service from “Good, I’m satisfied” to “Bad, I’m unsatisfied”.
The results are then used to calculate the Composite Customer Satisfaction Score. However, CSAT scores are usually expressed as a percentage: 100% representing total customer satisfaction, 0% total customer dissatisfaction.
If you’re curious about average CSATs in your industry and region, check Zendesk’s benchmark studies that may be beneficial.
How does it scale?
The possibility to rate conversations right in the middle of the interaction allows us to capitalize on excellent response time. However, it will also quickly punish us, if we are not fast enough.
Ratings given at the end of the interaction generally reflect the users’ more global appreciation of the service they experienced.
Figure out if you are interested in your enterprise’s loyal customers, or would you like to focus on the people who have just started to use your service or product? This will help you find the best place to target them.
For example, GetFeedback changed their CSAT survey strategy and started embedding satisfaction surveys directly into their messages when cases were closed. While they saw the CSAT score drop, they got useful qualitative insights for their support strategy as they grew.
Internal Quality Score
Internal quality score is the KPI of conversation reviews. Customer support conversation reviews are systematical internal evaluations of the interactions held between support team and customers. These reviews are done as self-reviews, manager-reviews or peer-reviews.
Think of it like CSAT, but instead of customers it’s your team providing internal feedback on the support efforts. Conversation reviews are one of the most simple, yet most powerful ways of improving the quality of your customer support.
For example, in a peer review cycle, each customer support team member (even the newest ones!) reviews a set of conversations from other agents and rates them based on the scorecard you’ve designed.
The scorecard consists of the rating categories that reflect the values that are important to your support organization, and the qualities of a successful support interaction. These make up the aggregate Internal Quality Score.
Here’s an example of a simple scoring system:
- Set up 3-5 categories for review that represent the most important qualities of your interactions (e.g. “tone”, “technical knowledge”, “use of help materials”).
- Calculate an internal review score per ticket with a binary scoring system. In each category, reviewers can give a “positive” or a “negative” rating. Each category then counts for ⅓ – ⅕ of the total score, depending on the number of categories you have.
- Calculate the total internal review score. For example, if a ticket receives 3 positive and 1 negative rating, the total score would be 75%.
The quality of your support interactions affects almost all relevant customer service metrics. By analyzing your interactions, you find gaps in your knowledge. That’s how you can improve your services.
How does it scale?
As you grow, it will quickly become impossible to keep track of every single support conversation. Instead of aiming to read all tickets, define a percentage of the total ticket volume that you review every month.
The percentage of reviewed conversations can eventually become rather small. For example, with 50 000 customer conversations every month, it may be difficult to review more than 5% of the total volume.
The most important thing when scaling your team is consistency. Focus on reviewing a sample that represents the total mass of conversations fairly. Also, make sure that reviews are distributed equally among the reviewers.
Oftentimes, larger companies hire dedicated training/QA specialists for this job. Without designated resources and tools for doing conversation reviews, this task can become very cumbersome.
This is where Klaus can help – Klaus is a tool that allows you to ditch spreadsheets, do conversation reviews seamlessly and improve the quality of your customer support.
This is the main metric when it comes to knowing how your support team is doing. For some teams, this reflects just the number of conversations in the inbox. For others, it combines all interactions coming in from social media, phone, and chat support.
Tracking conversation volume over a long period of time, e.g. years, can give you great insight into your support team trends. This will help you find your busiest periods and understand when you need to hire more staff.
How does it scale?
By analyzing how your customer support load is changing over time, you’ll understand when and how you need to hire. Knowing how much volume you get outside your business hours will help you decide when to start offering real-time support.
For example, if your volume spikes every Christmas, think about hiring seasonal support people or using a service provider like FCR. If the trend still continues to go upwards, it may be time to start hiring more full-time staff.
If you have a handle on how your volume normally trends, you can even start the hiring process in a low point of volume. This way you’ll have someone ready for when you really need them.
Cost Per Conversation
Cost Per Conversation (CPC) represents the total cost of operating your team, divided by the total number of conversations that you have across your support platforms. These costs include salaries, health insurance, and other benefits, equipment, and everything else that you need to have the support team running.
CPC = total team operating costs / total number of conversations
You can calculate CPC during a specific period of time – for example, across the span of a year or a month. Note that you should only count for the time that your agents are actually doing customer support tasks. That’s an important aspect if your team members spend some of their time on responsibilities other than support.
How does it scale?
As you get larger, you may need to find ways to automate some of the tickets and run a larger-scale support team easier. Knowing your CPC can help you determine if a new tool could help you save costs or make a huge impact on your bottom line.
CPC is especially useful when comparing different channels. For example, if email is cheaper than chat, you may want to scale it up in your offering.
You may also notice that your CPC is steadily rising, which means that it may be time to see if there is something about your strategy that you can change. However, those “costly” conversations might have some other positive impact, so pay attention to other metrics that align with it.
For example, maybe customers who get phone support convert to paid accounts at a higher rate? Make sure not to look at the cost strictly within the context of support, but the whole business.
Calculating your CPC can also help you understand if your improved documentation or ticket deflection have paid off. For example, Lance from Raven Tools noticed a clear change in it after updating and improving their documentation’s searchability.
First Contact Resolution rate
Across all forms of contact, First Contact Resolution (FCR) rate means solving the customer’s issue in full with your first response. Clients do not have to ask any more questions. Customers love getting their questions solved quickly.
Not all tickets are FCR-eligible. For example, sometimes customers don’t provide all the necessary information and it just isn’t possible for you to fix their problems on the first go. Non-FCR-eligible tickets should be tagged manually or with a service like Idiomatic.
However, with FCR-eligible tickets you can use the following formula from Groove to calculate your FCR rate:
FCR Rate = number of cases resolved on first contact / total number of FCR-eligible cases
How does it scale?
FCR should continue to remain the same as your company grows. If FCR is getting worse while your company grows, it may be a sign that your customer support representatives are rushing through a first response.
If your team is growing but your FCR is going down, there are a few things this might tell you about how your team is scaling:
- Your new employee onboarding and training is not as effective as it once was;
- You have more tickets than your team can effectively support, and the quality of their responses is starting to diminish;
- You’ve automated the easier tickets so that what remains is difficult to resolve in a single response – well done!
which of these is the case in your team? Go tackle it!
Time to First Response
First Response Time (FRT), also known as Time to First Response, is the metric that indicates how long your customers have to wait before they get a response to their inquiry.
It all boils down to everyone wanting to know that help is on the way, even if the resolution takes more time. Just like you would when calling 911.
Some leaders believe that customer reps should not reply to a ticket before they have an answer to their client’s problem. But that is often not true.
“… it’s indisputable — a speedy first reply results in higher customer satisfaction.” Anton de Young, Zendesk blog
If you’d like to learn more about customers’ expectation on FRT, here’s an infographic with some good stats.
How does it scale?
Zendesk has done a number of studies on customer satisfaction, and all of them come to the same conclusion: people love fast reply times. It’s the easiest way to under-promise and over-deliver to your customers.
If you allocate more resources towards your first response, CSAT goes up.
It is not about finding a direct answer right away, it’s about reassuring that a human has looked over what they’ve said.
To maintain a fast response time:
- Assign specific team members to individual platforms. For example, one agent could be responsible for Twitter, while another one manages the first response in your inbox.
- Utilize the round-robin ticket assignment: distribute cases to team members evenly as they come in. This will help you cherry-picking in your team and lowers FRT.
Eventually, when AI becomes smart enough, it can take on the responsibility of triaging the tickets and handling first responses.
Alternatively, a categorization machine-like Idiomatic’s can help you categorize cases. It will also identify problem areas to help you free up resources for quicker first responses.
Measure what’s most important to you
Don’t let the mire of available metrics get your team bogged down in the details. Take advantage of your customer service tools and the metrics that you already use. Then scale it.
Recognize that, when measuring team performance, quantitative data is not always king and sometimes qualitative data – in the shape of peer review or internal quality score – can be just as important for scaling your customer service.
Your metrics will grow with you as long as you know how to best track them and scale them for the needs of your team and company.
Follow us on Instagram for daily yoga tips for cats.