Your customer satisfaction scores look decent, but retention is low. So what’s missing? And how would you even know if you’re not tracking it?
According to Klaviyo's 2026 customer service research, 53% of brands track CSAT as their primary customer service metric.
Yet CSAT alone can’t tell you whether your customers find it easy to get help, whether they’d recommend your products, or whether your service interactions are actually driving revenue.
Now that 87% of service leaders view support as a core brand differentiator, according to our customer service research, it’s time to track metrics that prove impact. That means looking beyond operational efficiency to metrics that measure customer sentiment, effort, and loyalty.
Here are 7 essential customer service metrics that reveal what your customers really want, backed by insights from our customer service research, plus what you can do to improve them. (Hint: no metric exists in a vacuum, so there’s a lot of overlap between those strategies.)
1. Customer satisfaction (CSAT) score
Customer satisfaction (CSAT) score measures how satisfied a customer is with a product or service. You can measure CSAT with a simple survey question: “On a scale of 1–5, how satisfied are you with the [product/service] you received?”
These surveys are most accurate when you send them immediately after high-impact moments, like when a support ticket is resolved or after a purchase, return, or exchange.
Unlike broader sentiment metrics like net promoter score (more on this later), CSAT captures customer sentiment in the moment. For example, if a shopper gives a CSAT score of 5 out of 5 after ironing out a return with a customer service agent, that’s immediate validation that the interaction exceeded expectations.
On the flip side, low CSAT scores are early warning signals for immediate friction. They provide a quick opportunity to investigate what went wrong. Dig deeper by reviewing specific conversations, looking for patterns across tickets, or following up with the customer to understand what caused frustration.
How to improve CSAT scores
- Make CSAT scores actionable. Identify themes in common issues, like repeated questions about returns, delayed shipments, or confusing product details. Use these insights to train both your AI and human agents, and update the processes or on-site resources that are creating friction.
- Set up real-time CSAT tracking dashboards so your team can spot trends and respond quickly.
- Use AI-powered sentiment analysis on customer reviews to supplement CSAT scores and catch issues you might miss in survey responses alone.
2. First response time (FRT) and time to resolution (TTR)
First response time (FRT) measures how long it takes to provide an initial response to a customer inquiry. Time to resolution (TTR), meanwhile, tracks how long it takes to fully resolve their issue.
In recent years, customer expectations have accelerated thanks to AI and the “Amazon effect.” Shoppers expect instant updates, rapid responses, and near-immediate resolutions. FRT and TTR are customer service metrics that can help you meet these expectations.
According to our customer service research, only 35% of companies track response time and TTR. That means your support team may be able to gain a competitive edge if you make it a priority this year.
It’s important to remember, here, that speed matters, but not at the expense of customer service quality. Customers want fast replies that are also accurate and helpful. According to Klaviyo’s research, 25% of brands prioritize maintaining service quality even if response times slow during peak periods.
How to improve FRT and TTR
- Use an AI customer agent for instant first responses to common, straightforward questions. Customers get help right away, and your team has more time to focus immediately on inquiries that need a human touch.
- Adopt a helpdesk with AI-powered ticket routing and prioritization so the right human agent sees and handles urgent issues immediately. This empowers agents with certain kinds of expertise to shine and do what they do best.
- Bonus points if the helpdesk is omnichannel. That way, your agents don’t waste time toggling between different tabs and windows when a customer switches channels. They can view all threads of a conversation in a single workspace, no matter where the customer reaches out: web chat, email, text messaging, WhatsApp, or social.
- Review ticket management and performance metrics to find areas for improvement. You might find that certain types of tickets consistently take longer, or maybe handoffs between teams are slowing things down.
When apparel brand Ministry of Supply implemented a 24/7 AI customer agent that pulls from its storefront and FAQs to provide quick, accurate answers to customer questions, the agent handled 75% of “where is my order?” chat queries in a 60-day period.
3. Customer effort score (CES)
Customer effort score (CES) measures how much effort a customer has to exert to resolve their issue. You can measure CES with a one-question survey answered on a Likert scale, with emojis, with statements, etc., as long as the result translates to a number you can easily track.
CES is useful for evaluating whether your self-service experience is really helping or creating more friction for your customers. Still, only 25% of companies track CES, according to our customer service research.
But it’s important, because people don't always need to be wowed. Often, they just want things to be easy. Gartner research found that 96% of customers who encounter high-effort service interactions become more disloyal, compared to just 9% who have a low-effort experience.
How to improve CES
- Create a personalized, self-serve customer hub where customers can manage orders, find personalized product recommendations, chat with an AI customer agent for instant assistance, and get answers to FAQs—all without involving a human.
- Centralize customer data in a CRM with an embedded helpdesk. That way, your team doesn’t need to switch tabs or sift through information to understand customer’s history and needs, and customers don’t have to repeat themselves.
Home fragrance brand Happy Wax is a great example of how introducing a customer hub can reduce customer effort and deflect service tickets. The brand saw a 75% drop in customer support tickets related to tracking orders after implementing a self-serve customer hub, while also increasing repeat purchases.
4. First contact resolution (FCR) rate
First contact resolution (FCR) rate measures the percentage of customer issues resolved in the first interaction without follow-up. You can measure FCR by taking the number of customer issues resolved in a single interaction across channels within a specific time frame and dividing it by the total number of customer issues within that same time period.
A high FCR rate shows that your human agents have the knowledge, data access, and authority to efficiently solve problems, so there’s less need for follow-ups or repeat tickets.
This helps support run more smoothly, lower operating costs, and improve the quality of service experiences. But according to our customer service research, only 18% of brands currently track their FCR rate.
How to improve FCR
- Use AI customer agents that pull from your product and customer data to answer questions. When customers can get immediate and personalized support from AI, they may be less likely to escalate tickets to humans.
- With a CRM that centralizes data into unified customer profiles visible in your helpdesk, both AI and human agents can see what customers bought, how they’ve engaged over time, and where they’ve needed help before. With that context, service teams can respond instantly, personalize every interaction, and increase FCR rates.
- An omnichannel helpdesk that connects human agents to customers on the channels they prefer to use for customer service can help here, too. If a customer can resolve their issue via Instagram without switching to chat, for example, that can improve FCR rates over time.
5. Net promoter score (NPS)
Net promoter score (NPS) measures the likelihood someone will recommend your brand to others. Measure NPS with a one-question survey: “On a scale of 0–10, how likely are you to recommend [brand] to a friend?”
Responses fall into 3 groups:
- Detractors (0–6): unhappy customers who might actively discourage others from buying
- Passives (7–8): satisfied but unenthusiastic customers who are unlikely to make an effort to recommend you to others
- Promoters (9–10): loyal enthusiasts who will happily recommend you to others
These responses are calculated into a single NPS score, making it easy to benchmark over time. But according to our customer service research, only 13% of brands are tracking NPS.
How to improve NPS
- After support tickets are resolved, send automated NPS surveys to understand whether immediate service interactions can move the needle on recommendations. If you’re concerned about survey fatigue, you can temporarily pause other survey automations and run NPS surveys for a short time on a monthly, quarterly, or biannual basis.
- Add follow-up questions to understand why a customer gave a particular score. This can uncover problems like long wait times or unresolved issues that may drive customers away.
- Automatically create follow-up tickets for low scores from high-value customers, so your team can reach out proactively and potentially reverse some trends.
6. Customer retention and churn rate
Retention rate measures the percentage of customers who purchase again over a given period, while churn rate measures the percentage who don’t buy again.
Excellent customer service is a key driver of retention. After product quality and affordability, high-quality customer service is the top reason consumers stay loyal to a brand, according to Klaviyo’s 2025 Future of Consumer Marketing report.
But many brands don’t track retention or churn, either: our customer service research found that only 35% of companies actively monitor these metrics.
How to improve customer retention rates
- Use recency, frequency, monetary (RFM) analysis to segment loyal and lapsed customers, and proactively reach out with rewards or offers.
- Set up automated win-back flows triggered by customer service issues. For example, send a personal apology from your CEO or a discount when a high-value customer rates a support experience poorly.
- Create customer health scores that combine customer service metrics (like CSAT and response time) with purchase behavior (like order frequency and average order value) to get a complete picture of customer satisfaction.
7. Revenue or up-sells influenced by customer service interactions
Rather than seeing customer service as a cost center, brands are beginning to track the revenue it generates. According to our customer service research, 24% of brands currently track revenue or up-sells and cross-sells that are influenced by service interactions.
When customers contact support, they're engaged. It’s the perfect time for both AI and human agents to suggest relevant products, subscription upgrades, or complementary items if the interaction is going well.
How to improve revenue and up-sells from customer service
- You guessed it: a CRM is crucial here, too. When both AI and human agents have a complete view of each customer, including all browsing, preference, and purchase data, beside their ongoing conversations, they can offer relevant, personalized recommendations that are more likely to land and result in a purchase.
- Suggest relevant products, upgrades, or add-ons when customers ask questions about product features or shipping. AI customer agents can support this by offering personalized recommendations during these kinds of customer service interactions.
- Send automated follow-ups after resolving a customer issue, including offers tailored to their recent purchases or interests. This can turn a positive support experience into an opportunity for additional sales.
Improve customer service metrics with Klaviyo Service
Tracking customer service metrics only matters if you can act on them—and that requires integrated data.
When customer interactions, purchases, and support history live in separate systems, patterns are easy to miss and decisions get harder. Collecting all customer data in one, centralized place gives customer service teams the visibility they need to track the metrics to drive impact.
Klaviyo Service brings together the tools you need to track, analyze, and act on customer service metrics in one platform, with:
- Klaviyo Customer Hub: a personalized, self-serve, on-site destination where shoppers can manage orders, redeem offers, explore product recommendations, and get support
- K:AI (Klaviyo AI) Customer Agent: a 24/7 AI customer agent that’s trained on your storefront and customer data to answer questions, suggest relevant products, and resolve common issues across web chat, email, text messaging, and WhatsApp
- Klaviyo Helpdesk: an AI-powered helpdesk that unifies AI and human support in a single workspace across email, chat, SMS, WhatsApp, and social, helping teams respond faster with full customer context and more personalized interactions
When you give your teams full visibility into customer interactions and behavior, your customer support gets faster, more personal, and more effective—transforming customer service from a reactive function into a driver of loyalty and revenue.
Related content
