CSAT vs. NPS vs. CES: Key Differences and When to Use Each

CSAT, NPS, and CES are the three most common customer experience metrics, and each answers a different question. Customer satisfaction score (CSAT) measures how satisfied a customer is with a specific interaction. Net Promoter Score (NPS) measures how likely a customer is to recommend your business, which reflects the overall relationship. Customer effort score (CES) measures how easy it was for a customer to get something done, like resolving an issue or completing a purchase.

Most teams get the clearest picture by using them together: CSAT and CES at key moments in the customer experience, and NPS on a regular cadence to track loyalty over time.

Below, we'll compare how each metric works, when to use it, what counts as a good score, and how to connect all three to retention and to what customers actually say in their conversations with your team.

CSAT vs. NPS vs. CES at a glance

What is CSAT?

Customer satisfaction score (CSAT) measures how satisfied a customer is with a specific interaction, such as a support call, an onboarding session, or a purchase. It's a transactional metric: it tells you how one moment in the customer journey went, not how the customer feels about your business overall.

A CSAT survey usually asks one question right after the interaction, such as "How satisfied were you with the support you received today?" Customers answer on a 1 to 5 scale, where 1 is very unsatisfied and 5 is very satisfied.

How to calculate CSAT: count the satisfied responses (4s and 5s), divide by the total number of responses, and multiply by 100.

CSAT = (satisfied responses ÷ total responses) × 100

For example, if 100 customers respond and 82 choose 4 or 5, your CSAT score is 82%.

What is NPS?

Net Promoter Score (NPS) measures how likely customers are to recommend your business to someone else. Because it reflects the overall relationship rather than one interaction, NPS is a relational metric and a common indicator of loyalty.

The NPS survey asks, "On a scale of 0 to 10, how likely are you to recommend [company] to a friend or colleague?" Responses fall into three groups:

• Promoters (9–10): loyal customers likely to recommend you

• Passives (7–8): satisfied but not enthusiastic, and open to competitors

• Detractors (0–6): unhappy customers at higher risk of churning or sharing negative feedback

How to calculate NPS: subtract the percentage of detractors from the percentage of promoters. Passives count toward the total but not toward either group.

NPS = % promoters − % detractors

For example, if 55% of respondents are promoters and 15% are detractors, your NPS is 40. Scores range from -100 to +100.

Many companies run relational NPS surveys on a schedule, such as quarterly or twice a year. Some also run transactional NPS after specific milestones, like onboarding or a renewal, to see how those moments shape loyalty.

What is CES?

Customer effort score (CES) measures how easy or difficult it was for a customer to do something, like resolve a support issue, return a product, or set up an account. It focuses on friction: the extra steps, transfers, and repeated explanations that make an experience feel like work.

A CES survey usually asks, "How easy was it to resolve your issue today?" or asks customers to rate their agreement with a statement like "The company made it easy for me to handle my issue." Common scales run from 1 to 5 or 1 to 7, from very difficult to very easy.

How to calculate CES: there are two common methods. You can average all responses, or you can calculate the percentage of customers who chose the "easy" end of the scale (for example, 5, 6, or 7 on a 7-point scale). Whichever method you choose, use it consistently so you can compare results over time.

Key differences between CSAT, NPS, and CES

The biggest difference is scope. CSAT and CES are transactional: they measure a single interaction or task, close to when it happened. NPS is relational: it measures how customers feel about your business as a whole, built up over many interactions.

That difference shapes what each metric can tell you:

• CSAT shows whether a specific interaction met expectations. It's useful for evaluating support teams, channels, and individual agents, but a satisfied customer today can still leave next quarter.

• NPS shows the overall health of the relationship and helps you spot customers at risk of churning. It doesn't tell you which part of the experience to fix.

• CES shows where your processes create friction. It's narrower than CSAT, but it points directly at operational problems like repeat contacts, long hold times, and transfers.

Can you convert NPS to CSAT?

No. NPS and CSAT use different questions, scales, and calculations, and they measure different things. A customer can give a 5 on a CSAT survey after a good support call and still be a detractor because of pricing or product gaps. Track them side by side instead of trying to translate one into the other.

When to use CSAT, NPS, and CES

The right metric depends on the question you're trying to answer. Many teams use all three at different points in the customer lifecycle.

After a support call or chat: CSAT and CES

Use CSAT to measure customer satisfaction with the interaction itself, and CES to see whether getting help took more effort than it should have. Send both right after the conversation ends, while the experience is fresh.

On a regular schedule or ahead of renewal: NPS

Use NPS to track loyalty across your customer base over time. Quarterly or twice-yearly surveys can show trends without overloading customers. For B2B companies, an NPS survey ahead of a renewal can surface detractors while there's still time to address their concerns.

After onboarding or a self-service flow: CES

Use CES to find friction in processes customers complete on their own, like account setup, returns, or a help center search. High-effort responses point to the steps worth simplifying first.

After a product launch or major change: all three

CSAT shows how individual interactions are going, CES shows whether the change made things harder, and NPS shows whether it moved overall loyalty.

What is a good CSAT, NPS, or CES score?

A good score depends on your industry, your customers, and how you collect feedback. Published benchmarks are a useful starting point, but your own trend over time is usually the more reliable guide.

What is a good CSAT score?

A CSAT score between 70% and 85% is generally considered good, and scores above 85% are considered excellent, according to SurveyMonkey's CSAT benchmark guidance. Expectations vary by industry, so a score that's strong in one sector may be average in another.

What is a good NPS?

Any score above 0 means you have more promoters than detractors. Based on SurveyMonkey benchmark data from more than 150,000 organizations, the average NPS is 32 and the median is 44. The top 25% of organizations score 72 or higher, while the bottom 25% score 0 or lower.

NPS also varies widely by industry. Here are median scores and quartile ranges for a few sectors from the same SurveyMonkey data:

What is a good CES score?

CES benchmarks are less standardized than CSAT or NPS because companies use different questions and scales. Instead of comparing against an outside number, track your own CES over time and compare it across channels, teams, and processes.

The trend is worth watching closely. Gartner research found that customers who have a low-effort service experience have a 61% likelihood of staying with the company, compared with 37% after a high-effort experience.

How to track CSAT and NPS for customer retention

CSAT and NPS are most useful when they're tied to what customers actually do next, like renewing, buying again, or leaving. These steps can help turn survey scores into a retention signal:

1. Set a cadence for each metric. Trigger CSAT (and CES, if you use it) automatically after key interactions. Run NPS on a fixed schedule so you can compare results period over period.

2. Connect scores to customer records. Store each response with the customer's account, plan, tenure, and recent support history, usually in your CRM. A score without context is hard to act on.

3. Segment the results. Compare scores across new and long-term customers, plan tiers, channels, and teams. An average can hide a segment that's struggling.

4. Flag at-risk customers. Treat NPS detractors and low CSAT scores as early warnings. Route them to an account manager or customer success team for follow-up.

5. Ask why. Add one open-ended question, such as "What's the main reason for your score?" The answers often matter more for retention than the number itself.

6. Close the loop. Follow up with unhappy customers, fix the issue where you can, and let them know what changed.

7. Compare scores with retention outcomes. Check whether detractors and low-CSAT customers actually churn at higher rates. Over time, this shows which signals your team should act on first.

The limits of survey-based metrics

Surveys only capture the customers who choose to answer them. Response rates for post-interaction surveys tend to be low, and customers with very positive or very negative experiences are often the most likely to respond. That can skew your scores away from how most customers actually feel.

Surveys also lag. By the time a quarterly NPS result arrives, the experiences behind it may be weeks old. And even with an open-ended question, a score may not tell you exactly what happened in the conversation that produced it.

Measuring satisfaction from conversations, not just surveys

The most detailed customer feedback many businesses have isn't in a survey. It's in their everyday customer conversations: what customers ask for, where they get frustrated, and how they react when an issue is resolved.

When every call and message happens in one place, those conversations can become a reliable source of conversation intelligence. With Dialpad, contact center teams can:

• Estimate satisfaction without a survey. AI CSAT uses call transcripts between agents and customers to detect CSAT scores across eligible calls, so you aren't relying only on the customers who fill out a survey. AI CSAT explanations go a step further by identifying the reasons behind each score.

• Track sentiment as it happens. Sentiment analysis shows how customer sentiment shifts during calls, so supervisors can step in while there's still time to help.

• Find patterns across conversations. Conversation intelligence surfaces recurring topics, questions, and complaints, so you can see what's driving your scores.

Surveys still matter. CSAT, NPS, and CES give you a direct, comparable read on how customers feel. Pairing them with conversation data gives you a deeper understanding of your customers, helps your team turn interactions into operational insight, and shows which changes are most likely to improve CSAT scores in your call center.

Measuring customer experience when AI agents handle interactions

As more businesses use AI agents to answer customer questions, measuring satisfaction carefully becomes even more important. A high deflection rate shows that an AI agent handled a conversation without a human. It doesn't show whether the customer's issue was actually resolved, or how the customer felt about it.

When AI agents handle part of your customer experience, consider tracking:

• Resolution, not just deflection: whether the customer got what they needed, or contacted you again about the same issue

• Sentiment and satisfaction: how customers feel about AI-handled interactions compared with conversations with human agents

• Effort at handoff: whether customers have to repeat themselves when an AI agent escalates to a person

Dialpad Resolve, AI agents for voice and digital interactions, can resolve routine requests and hand more complex conversations to human agents along with context such as the reason for the contact and the steps already attempted. Because Dialpad's UCaaS, CCaaS, and AI agents run on one platform, AI-handled and human-handled interactions can be reviewed side by side when you measure customer experience.

Other metrics to pair with CSAT, NPS, and CES

CSAT, NPS, and CES tell you how customers feel. Operational and business metrics show what's behind those feelings and what they cost. Common pairings include first contact resolution, which often tracks closely with effort; churn rate, which shows whether low scores turn into lost customers; customer lifetime value; and customer sentiment from conversations.

The right mix depends on your goals, and a broader set of customer satisfaction metrics and customer engagement metrics can round out the picture. In a contact center, these sit alongside operational call center KPIs like average handle time and service level.

No single metric tells the whole story, so the most useful approach combines survey scores, conversation data, and business outcomes to help your team make better decisions about the customer experience. That's the idea behind Dialpad's AI platform for customer experience, which brings calling, messaging, contact center, and AI agents together so customer conversations can inform those decisions.