The Alphabet Soup of Customer Service Metrics

The Alphabet Soup of Customer Service Metrics: CSAT, NPS & CES

When my grandma points to examples of great customer service, it’s usually because the person she interacted with was pleasant and polite, or maybe threw in a little extra something in the bag. For her, these are the hallmarks of great service. She spent her childhood in an age where gas station attendants were dressed as well as modern-day office workers.

This got me thinking: “how do we measure customer service excellence?” It’s easy to relay an anecdote about great service, but how do you extract data from an anecdote? How can you measure an interaction that is inherently personal? Fortunately, we have a few metrics we can use to understand where our experience is succeeding and failing.

Understanding What Great Service is According to the Numbers

The survey metrics that you assign to agent scores or use to quantify customer happiness are only useful if you understand how they impact your business. Don’t miss an opportunity to improve your customer experience and make a costly long-term mistake. Instead, leverage the data you’ve collected to drive smarter-decision making (according to the numbers & not gut). However, the moment you start making adjustments to your team, process, or policy for the sake of improving the score (instead of your service), STOP.

I’ve seen many teams implement changes that “improved” one key performance indicator (KPI), but at the cost of creating a worse customer experience. For example, some team leads remove an effective support channel because it lowered the First Contact Resolution (FCR) rate even though customer satisfaction was trending high. The team’s turnaround time was under 5 minutes, and the cost to resolve an issue was the lowest of all support channels.

Never use numbers to justify a change for the sake of single metric. Always think about the end-user experience first, and filter all decisions through your higher-level vision and goals. KPIs should be deduced from the vision, not the other way around.

Remember the ultimate goal is a great customer experience.

Obstacles All Too Common

Breaking Down the Acronyms: How to Choose the Right Service Metric for Your Organization

CSAT: Customer Satisfaction  Score

This is usually a score that is represented as a percentage (0% – 100%) and often asked in some variation of the question: “On a scale of 1–10, how satisfied were you with your most recent support interaction”.

Managers often assume that the more satisfied customers indicates a strong correlation to a customer’s loyalty. However, CSAT scores are often too subjective because a customer’s answer is biased based on the outcomes or policies they may not like. This metric only seeks to understand the customer’s satisfaction with the most recent interaction, and ignores the process of getting to the solution. It’s a poor indicator of future loyalty and can misrepresent the health of your relationship with the customer.

NPS: Net Promoter Score

 Quite easily the most popular new metric for measuring customer loyalty. Phrased as the question: “Based on X experience, how likely are you to recommend Brand Y”.

Respondents then select a number from 1–10, with 9–10 classified as “Promoter”, or someone that is likely to tell others about their positive experience, 7–8 as “Neutral”, and 1–6 as “Detractor”, someone that will actively tell others to avoid your brand. The stat about how likely someone is to tell their friends about a negative experience vs. a positive one is bandied about quite often when discussing NPS.

NPS is a good measure of how well your team is performing, but the focus is on outcomes instead of the process, and even then this doesn’t necessarily reflect whether or not someone will actually refer their friends based on their most recent support experience.

CES: Customer Effort Score 


Simply stated, CES reflects how hard a customer has to work to resolve their issue. This can be through any channel, be it phone, email, self-service, etc. CES is measured on a scale of 1–10, or using the standard “Agree/Disagree” spectrum.

Research from Harvard Business Review shows that ‘delighting’ customers doesn’t build customer loyalty, however simplifying the process to resolve an issue positively impacts customer loyalty. It’s in your best interest to remove as many barriers as possible when a customer attempts to resolve an issue. Using this knowledge you can improve not only your customer service strategy, but also reduce service costs and lower the customer churn rate.

Which metric should I use?

Well, of course that depends on what you’re doing, and more importantly, your service operations goals. Most companies I’ve worked with tend to use NPS and CSAT in the early stages of building a support team; I feel this is backwards if your goal is to build customer loyalty or at the very least avoid losing customer loyalty.


CES should be the most important metric early on as it will give you the most meaningful feedback to start building and scaling tools with effective processes. Harvard Business Review notes that “the superior performance of CES in the service environment derives from two factors: its ability to capture customer impressions at the transactional level and its ability to capture negative experiences as well as positive ones.”

CES will help you start lean and execute efficiently, and grow from there because more often than not additional headcount is rarely the best solution to support challenges. When a team is small, you tend to hire only agents that will provide quality support, so naturally CSAT/NPS will be high. Where you’re failing (without knowing it) is that this level of individualized attention isn’t scalable without considerable resources, nor is it an accurate representation of customer loyalty and satisfaction.

You know your situation best. For example maybe you already have great self-service tools and the challenges you’re facing more closely align with general customer happiness with your brand or company. In this case, you might want to promote efforts to improve NPS or CSAT scores.

It’s all dependent on the challenges you face. The key is to understand the benefits along with the pitfalls of each metric scale. NPS and CSAT will be the most subjective because the score is impacted by factors outside of the support agent’s control. CES will tell you if your service is being delivered as intended and identify whether the proper infrastructure is in place to sustain an effective help center.

It may seem that I’m biased towards CES, and well, I am. Removing obstacles to allow your customers to resolve their issues is the best way to ensure they return to your business again and again.

Similar Posts