NPS vs CSAT vs CES: Which Customer Metric Should You Track?

Three acronyms get thrown around any time someone talks about measuring customer feedback: NPS, CSAT, and CES. They sound interchangeable. They are not. Each one asks a different question, captures a different moment, and tells you a different thing about your store.

This guide breaks down the NPS vs CSAT vs CES debate in plain terms, shows you when each metric earns its place, and explains how to actually run them on Shopify after a purchase. Spoiler: most stores should start with exactly one of them.

What each metric actually measures

NPS (Net Promoter Score)

NPS asks one question: "How likely are you to recommend us to a friend or colleague?" Customers answer on a 0 to 10 scale. You then bucket the responses:

• Promoters (9-10): loyal, will refer you
• Passives (7-8): satisfied but unenthusiastic
• Detractors (0-6): unhappy, may churn or warn others

Your score is the percentage of promoters minus the percentage of detractors, which lands somewhere between -100 and +100. NPS measures loyalty and overall sentiment, not any single interaction. It is a relationship metric.

CSAT (Customer Satisfaction Score)

CSAT asks how satisfied someone was with a specific thing: an order, a support reply, a delivery. The common format is a 1 to 5 star rating, sometimes 1 to 7. You calculate CSAT as the percentage of responses that land in the top buckets (4-5 stars).

CSAT measures satisfaction with one transaction or touchpoint. It is narrow and immediate. Ask it right after the thing happens and the number means something.

CES (Customer Effort Score)

CES asks how hard a customer had to work: "How easy was it to place your order?" or "How easy was it to get your problem resolved?" Usually a 1 to 7 scale from "very difficult" to "very easy."

CES measures friction. It is most useful for support interactions and checkout flows, where effort predicts churn better than satisfaction does. A customer can be satisfied with the outcome and still annoyed by how long it took.

NPS vs CSAT vs CES at a glance

	NPS	CSAT	CES
The question	Would you recommend us?	How satisfied were you?	How easy was it?
Scale	0-10	1-5 (stars)	1-7
What it tracks	Loyalty, relationship	Satisfaction with one touchpoint	Friction / effort
Best moment to ask	Post-delivery, post-purchase	Right after the interaction	Right after a task or support contact
Strength	Predicts referrals and retention	Pinpoints what people liked or didn't	Flags where customers struggle
Weakness	Vague on the "why"	Inflated by polite high ratings	Narrow, only useful at friction points

In OrderSurvey, NPS is the 0-10 question type and CSAT is the 1-5 star rating. CES doesn't have a dedicated widget, but you can build it with a single-select or dropdown question scaled "very difficult" to "very easy," so all three are covered with the available question types.

When to use which

Pick the metric that matches the decision you are trying to make.

Reach for NPS when:

• You want a single north-star number to track loyalty over time.
• You care about word-of-mouth and referrals.
• You want to flag unhappy customers early so you can intervene.

NPS shines as an ongoing pulse. Ask it a week or two after delivery, when someone has actually used the product, and the score reflects the whole experience rather than the unboxing high.

Reach for CSAT when:

• You want to grade a specific moment: the checkout, the packaging, the support reply.
• You're testing a change (new carrier, new packaging) and need a before/after read.
• You want quick, high-volume feedback that is easy for customers to give.

CSAT has the highest response rates of the three because a star rating costs almost no effort. That makes it the friendliest first question on a thank-you page.

Reach for CES when:

• You're debugging a process, like a clunky checkout or a slow return flow.
• Support is a big part of your customer experience.
• You suspect friction, not dissatisfaction, is driving people away.

CES is the most specialized of the three. Most ecommerce stores don't need it as a standing metric. They reach for it when something specific feels broken.

How to run these post-purchase on Shopify

The metric is only half the job. Where and when you ask decides whether anyone answers. OrderSurvey runs surveys on three native surfaces:

1. The thank-you page, right after checkout completes. Best for asking about the buying experience itself: CSAT on the order, or a CES on checkout ease.
2. The order status page, which customers revisit while tracking shipping. A good spot for NPS once the product has likely arrived.
3. Shopify POS, for in-store, post-sale feedback at the counter. See the POS survey guide for setup.

Because OrderSurvey is built on Shopify's native checkout, customer-account, and POS UI extensions, the survey renders inside the page without bolting on a script that drags down load time. It also means the app doesn't request broad data-access scopes, which matters if you'd rather not hand another app the keys to your customer data.

A practical sequence for a typical store:

• Thank-you page: one CSAT star rating on the purchase experience. Low effort, high response.
• Order status / a few days later: one NPS question once the product is in hand.
• Follow-up logic: if NPS lands at 0-6, branch to a short text box asking what went wrong. If it lands at 9-10, ask permission to share their words as a review.

That follow-up branch uses conditional branching, so detractors and promoters see different next questions instead of one generic form. For NPS specifically, the deeper mechanics live in our Shopify NPS guide.

Closing the loop

A score you never act on is a vanity metric. Two features make the data operational:

• Low-score alerts: when an NPS response comes in at or below a threshold you set, OrderSurvey can fire an alert (for example, to a Slack webhook) so someone can reach out while the order is still fresh.
• CSV export: pull every response out to slice by product, order value, or date and feed your own reporting.

Set the alert threshold to match the metric. For NPS, 6 or below catches detractors. For CSAT, treat 1-2 stars as the alert trigger.

Why most stores should start with one

Here's the part people skip. You do not need all three. Running NPS, CSAT, and CES at once usually produces three half-answered surveys and three numbers nobody trusts.

Survey fatigue is real. Every extra question lowers your completion rate. If you ask a customer for an NPS score, a CSAT rating, and a CES score in the same breath, most will bail before the end, and your data skews toward the handful of people patient enough to finish.

So pick one to start:

• Default to CSAT if you're new to surveys. It's the lowest-friction question, it gets the highest response, and it gives you something concrete to fix this week.
• Default to NPS if you care most about retention and referrals and want one trend line to watch over quarters.
• Add CES only when you have a specific friction problem to diagnose.

Get one metric running cleanly, build the habit of reading and acting on the responses, and then layer in a second question once you know your first one works. A single well-placed question that people actually answer beats a battery of metrics that they don't.

If you're deciding where that first question should live, the thank-you page vs order status page breakdown covers the tradeoffs. And whenever you're ready to put a metric in front of real buyers, install OrderSurvey and ship one question this week. You can run it free up to 100 responses every 30 days before deciding it's worth more.