Managing a customer support team without tracking key performance indicators (KPIs) is like taking a road trip in a foreign country without a GPS: confusing, stressful, and inefficient.
Monitoring customer service KPIs gives CX leaders a clear view of agent performance and the effectiveness of their coaching processes—both of which impact the company’s bottom line. You might be familiar with some foundational KPIs, such as CSAT. But did you know there’s another subset of advanced metrics you can track to level up your customers’ experience?
In this article, we’ll discuss two tiers of customer service KPIs:
Let’s get started.
This first tier of KPIs gives customer service leaders and stakeholders a valuable look behind the scenes at how well agents interact with customers and where there are areas for improvement. These are standard metrics every team needs to monitor.
The first three KPIs measure agent performance directly, and the last two evaluate customers’ perceptions of an agent or the company in general.
QA scores offer a powerful snapshot of how agents perform relative to the company’s standards for quality interactions. This is almost always the most important customer service KPI because it provides actionable data.
In other words, it indicates why—not just if—agents perform up to par.
How to Calculate QA Scores
QA managers calculate scores with a QA scorecard, which can contain questions such as:
QA managers assign points to agents for each question, and an agent’s score is the percentage of total available points they earn on their scorecard.
So, if an agent is consistently scoring low on a question about friendliness, a manager can focus future trainings on greeting techniques or using approved brand language.
QA scores usually fall between 75-90% based on four random conversation reviews per week.
Average Handle Time indicates the average amount of time it takes for a support agent to close a ticket. That includes hold time, the duration of the interaction, and any necessary follow-ups.
Nearly 60% of people are frustrated with long hold times, according to RingCentral. Accordingly, you should keep tabs on agents’ AHT to better understand which agents need to improve their efficiency.
How to Calculate AHT
AHT = (total call time + total hold time + follow-up time) / total number of calls
First Contact Resolution indicates the percentage of customer issues that are resolved on the first interaction.
To understand the importance of FCR, put yourself in the customer’s shoes: reaching out multiple times to customer support to get a question answered is a major pain. When agents resolve issues on the first contact, it reduces friction with the customer and may even help other KPIs, such as CSAT.
How to Calculate FCR
FCR = (number of tickets solved on first contact / total number of support tickets)
FCR varies by industry, but the average is 74%, according to research from MetricNet.
CSAT scores indicate how satisfied a customer is with a product or service. CSAT provides quick, easy-to-interpret data on customer sentiment. But the downside of this KPI is that it can’t tell you why a customer is satisfied or so furious they never want to do business with you again. In other words, it creates a CX blindspot.
How to Calculate CSAT
CSAT is measured through direct customer feedback with a question like:
On a scale of 1-5, how would you rate your overall experience with the product/service you received today?
To calculate your CSAT score, divide the number of positive responses (4 or 5, if you’re using a scale of 1-5) by the total number of responses, then multiply by 100. Here’s a quick example:
Let’s assume you have 60 total responses, and 45 of them are positive.
45 / 60 = 0.75 x 100 = 75
CSAT scores vary by industry, but 80 is a great benchmark to aim for.
Net Promoter Score gauges the likelihood of a customer recommending a company’s product or service. NPS isn’t the sole responsibility of customer support teams, but keeping track of NPS for individual agents can help CX teams draw a correlation between customer service and a customer’s willingness to promote the business to their family, friends, or co-workers.
How to Calculate NPS
NPS requires a bit more number-crunching than the previous metrics.
First, send customers the standard NPS question: “On a scale of 1-10, how likely are you to recommend us to a friend/family/colleague?”
Next, group respondents into three categories:
Last, calculate the percentage of respondents in each category and subtract the percentage of Detractors from the percentage of Promoters. That number is your Net Promoter Score.
Here’s a quick example:
[60% Promoters] - [30% Detractors] = +30 NPS
NPS scores greater than 0 indicate that your audience is more loyal than not, while scores greater than 80 are considered world-class.
Now that we’ve covered the essential KPIs, let’s explore some metrics that advanced CX teams can track.
Once you have a firm grasp on the first tier of KPIs, you can take the next step to audit your processes and unlock insights to improve them. Team that track Tier Two KPIs are usually more established, have more agents, and have specialized roles such as a QA analyst or training lead.
The following KPIs are often overlooked, so teams that dig into them can get an edge over their competitors and take their customers’ experiences to new heights.
As its name implies, ticket review volume represents how many customer support tickets a CX team reviews over a certain period of time.
One of the main reasons CX teams track this KPI is that tells leadership how effective their QA process is and highlights areas for optimization. For example, Monday.com set a goal to increase their ticket review volume by 48% over the course of three months, and as a result of the increased insights they gleaned, reduced their AHT by 30%.
An alignment score is a percentage that shows how QA managers’ scoring stacks up against an internal benchmark. This is known as “Grading the Grader.”
In this workflow, a random sample of graded tickets is sent to a Senior QA Grader (or Benchmark Grader) who sets the standard on how grading should be done.
The Benchmark Grader grades the ticket without seeing the original score to see how well-aligned the Original Grader was with the Benchmark Grader. If there’s a significant discrepancy, the two parties meet to resolve the issue instead of letting the agent get stuck in the middle.
Alignment scores are important to keep your graders consistent. It’s more than just checking if the grader has given the same final score as the Benchmark Grader—it zeroes in on individual parts of the scorecard to ensure alignment on every section.
How to Calculate Alignment Score
To calculate Alignment Score, a QA grader is assessed on how their grades stack up against the Benchmark Grader for every question (or criterion) on the QA scorecard.
Then, using the weight of each criterion on the rubric (if the criterion makes up 15% of the rubric, it’s given a weight of 15%), calculate the grader’s Alignment Score by adding the weighted scores of each criterion, up to a total of 100%. The closer to 100%, the better.
If that’s too much mental math, MaestroQA handles all the calculations for you—just run your Grade the Grader workflows every month to keep everyone aligned.
For a more detailed example of how to calculate Alignment Score, check out this spreadsheet.
High-performing CX teams track agent engagement, which is the percentage of graded customer support tickets reviewed by a particular agent.
Grading and compiling QA scores is helpful, but knowing how dialed in an agent is to their own development lets you know which ones are self-driven learners and which ones might need some prodding.
How to Calculate Agent Engagement
The equation is easy:
Agent Engagement = (Total Number of Graded Tickets / Number of Graded Tickets Reviewed by Agent)
This KPI indicates how often an agent disagrees with the QA scores they receive. If agents continually appeal their scores, that can be a sign of a mismatch between the team's CX goals and how those goals (and the corresponding customer service training and policies) are perceived by the agent.
Once teams track this KPI, they can set a target (for example, 2-3%), and if the ticket appeal percentage exceeds that target, they know to review their CX or QA policies and scorecards with agents to understand why that gap exists and how to close it.
How to Calculate Ticket Appeal Percentage
To determine your ticket appeal percentage, simply divide the number of appealed tickets by the total number of graded tickets, then multiply by 100.
Customer service is both an art and a science. It requires masterful people skills, as well as constant digging into the data behind interactions (and setting goals along the way) to turn bland customer experiences into remarkable ones.
Ready to see how data can level up your customer service? Get a free demo of MaestroQA today.