When you talk about measuring customer experience and satisfaction, three metrics inevitably come up as THE ones to use: Net Promoter Score (NPS) vs Customer Satisfaction Score (CSAT) vs Customer Effort Score (CES).

So, which is best for you?

While NPS, CSAT, and CES all garner feedback about how a customer feels about your company, products, and services, there are nuances that make one better than the other depending on your use case and business goals.

Whether you’re starting your first program, or are already a customer experience veteran, it’s always worthwhile to take stock of and experiment with the metrics at your disposal.

“You might be tempted to default to whatever metric you tracked at a previous company, accepting that as the single source of truth, but the fact is that no one experience is exactly the same as another. Having an arsenal of these customer satisfaction metrics makes sure folks aren’t trying to define an experience with a metric that just isn’t a great fit.”

— Sean Mancillas, Customer Concierge at Delighted

In this article, we’ll explore their differences to help you understand how to use each for measuring your customer experience.

First, let’s talk about how each of these metrics map to your goals.

The purpose of NPS, CSAT, and CES

When you start your customer experience program, the first step is to figure out what you’re trying to accomplish through the data gathered in your customer experience study. There are generally three types of studies:

1. Relationship studies, where you evaluate the quality of the overall relationship a customer has with you. The goal is to monitor how your customer base feels about your company over time as both your business and the market evolve. Insights gathered inform strategies that improve customer loyalty, and subsequently revenue through increased customer retention, reduced churn, and word of mouth.

2. Transactional studies, where you measure how well your company helps customers complete certain tasks, whether that’s with your product or your support team. The goal is to gain specific, actionable feedback to improve certain products, processes, and services.

3. Journey-based studies, where you map and evaluate all of the interactions your target audience (new or existing) has when they do business with you, with the goal of supporting them better each step of the way. Customer journey-based studies are the most sophisticated of the bunch, and often require both relationship and transactional surveys, in addition to user and market research.

Overall, NPS is primarily a relationship study metric (though it can also be leveraged for transactional studies — more on this later). CSAT can be either a relationship or transactional study metric, and CES is purely a transactional metric. You can get the full low-down on relationship vs transactional surveys here.

When you take a closer look at the single-question survey used to calculate each metric, you’ll see what makes these studies relational vs transactional.

NPS relationship survey question

On a scale from 0 to 10, how likely are you to recommend [company] to a [friend or colleague]?

What makes this a relationship survey is the fact that:

  1. The question isn’t specific to any particular event or transaction.

  2. The question is scoped to the brand overall, which means that if you survey your customers on a regular basis, you’ll be able to monitor customer sentiment toward your brand over time.

To make the NPS survey question transactional, you need to modify it a bit more: On a scale from 0 to 10, how likely are you to recommend [company] to a [friend or colleague] based on [your most recent interaction]?

Note that whether you’re using the relationship or transactional version of the NPS survey, you’re still asking customers if they’ll recommend your company. This requires taking their entire experience with you into account.

When to send NPS surveys: For new customers, send out your initial survey after they’ve had time to fully experience your product or service. From then on, send your survey out on a regular cadence (quarterly or biannually) to monitor evolving customer sentiment.

For more in-depth guidelines by industry, check out our post on when to send your NPS survey.

NPS score calculation: Your NPS is expressed on a scale of -100 to 100. Calculate NPS by subtracting the percentage of promoters (people who rate you 9 or 10) from the percentage of detractors (people who rate you 6 or lower).

CSAT relationship/transactional survey question

On a scale from 1 to 5, how satisfied were you with [company/interaction]?

By swapping out what you ask your customer to rate — your company, or a specific interaction, you can use your CSAT survey for either a relationship or transactional study.

In addition to adjusting copy, the timing of your survey can also help determine relationship vs transactional CSAT (i.e. surveys triggered twice a year as opposed to surveys triggered post-transaction).

When to send CSAT surveys: Transactional CSAT surveys should be timed for recency, depending on what you’re evaluating. For sales, onboarding, or support interactions, ask for CSAT feedback immediately. For product feedback, time the survey so that you’ve given the customer enough time to fully experience your product.

Relationship CSAT surveys would be sent regularly over longer periods of time, similar to NPS surveys.

CSAT score calculation: Your CSAT score is the percentage of respondents who are “4 – satisfied” and “5 – very satisfied.”

CES transactional survey question

On a scale of 1 to 5, rate how easy [company/person] made it for you to [accomplish a task].

Customer effort score surveys emphasize a specific task or interaction in the question, which makes it perfect for transactional studies. CES has also been correlated to customer loyalty.

When to send CES surveys: Similar to CSAT, you should ask customers to evaluate an interaction immediately after it happens. This could be just after a purchase, or after the resolution of a service ticket.

CES score calculation: Your CES score is the percentage of respondents who “4 – agree” and “5 – strongly agree.” Since NPS and CSAT can both be used for the relationship and transactional use case, let’s compare them first.

CSAT vs NPS for measuring customer experience

The reason NPS and CSAT tend to be used for relationship studies is because both require a customer to evaluate their experience with your overall company in the rating.

  • CSAT relationship question: How satisfied are you with [company]?
  • NPS relationship question: How likely are you to recommend [company] to a friend?

The main difference between CSAT and NPS is the built-in connotation of the words “satisfaction” and “recommendation.” Satisfaction tends to be a more short-lived sentiment, while recommendation tends to be harder won. In short, just because someone is satisfied, doesn’t mean they’d recommend you to a friend.

Because of that, CSAT is thought of as a short-term evaluation of customer happiness useful for targeted, iterative improvements, while NPS is more of a long-term indication of loyalty.

Now, let’s consider the transactional version of these two questions:

  • CSAT transactional question: How satisfied are you with [interaction]?
  • NPS transactional question: How likely are you to recommend [company] to a friend based on [your most recent interaction]?

The CSAT transactional question is still pretty straightforward. Customers can easily rate any interaction, whether it’s with a customer support agent or a particular product feature, in isolation of how they feel about your company overall.

On the other hand, the transactional NPS question could put some people in a pickle. For instance, let’s say you had to ask the cable guy to come fix something, and your cable service provider asked: “How likely are you to recommend us based on your cable guy?

Your cable guy could have provided a great experience, but maybe you dislike your cable company overall — there’s no getting around the spotty service, and you feel the quality doesn’t justify the prices. How would you answer? That type of hangup makes it difficult for customers to answer consistently, leading to a less reliable transactional metric.

CSAT pros and cons

CSAT is popular because of its versatility and widespread usage across industries and verticals. While it’s always best to benchmark your scores internally, you could easily benchmark yourself against competitors as well to help make a case to stakeholders for starting a customer experience program.

You can also easily adopt CSAT for both relationship and transactional studies, and it would be easy for everyone in the company to understand. Because the question format is so versatile, it is extremely effective for tracking how your customer feels through the entire customer lifecycle. If satisfaction drops at any touchpoint, you will have the insights necessary to follow up with the customer and close the feedback loop.

Finally, CSAT surveys give you the option to swap out the numeric rating scale for visual rating scales, like stars or smileys, which might be more on-brand for an engaging survey experience.

The con of CSAT score is that satisfaction isn’t necessarily correlated to customer loyalty. Satisfaction has more of a fleeting connotation — a customer could be satisfied with their last interaction, which leads to a better impression of your brand overall, but doesn’t necessarily mean they’ll recommend you to a friend.

NPS pros and cons

Since Bain & Company noted NPS as an indicator of future business growth due to correlation to customer loyalty, NPS has been widely adopted internationally by some of the world’s largest brands. If you use NPS, you’ll be able to benchmark against yourself and competitors using the standard relationship NPS study, which is helpful for winning buy-in from key company stakeholders. (You can learn more about what a good NPS score is here.)

In addition, the NPS scoring method segments your customers into 3 groups: promoters (ratings of 9 or 10), passives (ratings of 7 or 8), and detractors (ratings of 6 or lower), so you can tailor campaigns and solutions to win and maintain the loyalty of each.

A criticism of NPS has been that it can be difficult to get actionable feedback and identify areas to improve, since the rating applies to the entire customer experience at a high level.

While you can use NPS for transactional studies, the results may be inconclusive. To go back to our earlier example, you could love the cable guy, but dislike the company overall and still not recommend it. Or, you could have hated the cable guy, but still recommend the company because you think that one-off interaction isn’t all that important to your evaluation.

This makes it a bit tricky to get conclusive data from a standalone transactional NPS (tNPS) survey: while you may be able to tell how important a particular interaction is to the sentiment towards your company as a whole, you may have a hard time understanding just how happy the customer was with that isolated interaction.

That’s why it’s so important to ensure your customer feedback platform asks customers to explain why they chose a certain rating, and makes it easy for you to analyze those open-ended comments. These NPS case studies show how you can leverage the scores and feedback to improve your business.

So why might you choose CSAT vs NPS, or vice versa?

Choosing CSAT vs NPS boils down to whether you’re evaluating customer happiness with your products and services in the short term, or if you’re trying to measure long-term brand loyalty.

Oftentimes, brands will use both: NPS to keep an eye on the overall quality of the experience, and CSAT for more targeted service and product feedback.

Now that we’ve taken a look at CSAT vs NPS, let’s take a closer look at CES, which has the most specific use case.

NPS vs CES for indicating customer loyalty and customer experience improvement

Customer Effort Score made headlines when a Gartner study found that “Effort is the stronger driver of customer loyalty,” which put it in contention with NPS. However, CES still hasn’t taken off the way NPS has, making it difficult to use as a competitive benchmark.

CES explores how much effort a customer feels they had to put in to resolve an issue, and is usually used to help customer service improve resolution times and provide higher-quality experiences.

It is a meaningful metric mainly due to how badly one negative customer service experience can taint the overall brand interaction. For example, “Ninety-four percent of customers with low-effort interactions intend to repurchase compared with 4% of those experiencing high effort.”

Because of that, studies have found that the CES scores tend to correlate with the NPS score when it comes to measuring customer loyalty.

Still, does that mean you should only use one metric or the other?

CES pros and cons

In the case of CES, the hyper-focused question is a double-edged sword. It’s great for getting actionable feedback for certain use cases, but its scope is implicitly limited.

After all, ease isn’t a factor in everything. Once you’ve addressed the low-hanging improvements that can be made to customer service, purchase, or user experience interactions, there’s still the question of the overall experience to discuss.

So why might you choose NPS vs CES, or vice versa?

If you’re focusing on improving customer service metrics, CES surveys are a great way to help reduce time to resolution and customer friction. However, if you need a broader evaluation of your overall customer experience and an easy way to segment your customer base according to sentiment, NPS is the way to go. Or, you can always use both.

For example, we decided to use email and in-app NPS surveys alongside CES customer service surveys to monitor and improve the Delighted experience, though not before trying out CSAT surveys for support feedback as well.

One of our Customer Concierge team members, Sean Mancillas, had this to say about selecting CES for measuring customer sentiment in support interactions:

“We had noticed resolution times ticking upwards over the last 12 months. With experiments in CSAT and NPS, we never really isolated the key drivers that were causing these support conversations to trend longer and longer. That’s where CES came in. When we became aligned on measuring and improving ease-of-experience via CES, specifically when customers interacted with our Concierge team, we were able to pinpoint and correct help docs, walkthroughs, and other support resources that made it difficult for customers to get the answers they were after. CES helped us get out of the way between our customers and the answers they are after.”

Since not every experience, industry, or business is the same, experimentation can always help you narrow down what metrics make the most sense for you.

In review, NPS, CSAT, or CES?

As you’ve gathered, the question should instead be: What’s the goal of your customer experience strategy?

If you are going through an overhaul of your products and services, asking customers for CSAT feedback every step of the way can help you decide in real time whether the changes are a good idea. If you are looking to understand customer loyalty and how your brand is perceived overall, use NPS. If you want to isolate and improve areas of customer friction, CES is the ticket.

Of course, using a combination of these methods will give you the most comprehensive overview of customer sentiment. Sometimes, you may also need to experiment with multiple metrics to see which result in the most insightful, actionable feedback for your business.

No matter which metric you choose to track, as long as you use your customer feedback insights to close the loop and effect change, you will be well on your way to delighting your customers.

Ready to start measuring customer satisfaction? You can use Delighted to send all of the survey types we’ve discussed: NPS, CSAT, CES, Smileys, 5-Star, and even Thumbs. Sign up now and send your first 250 customer experience surveys for free!