What Is a CSAT Survey?
Learn what a CSAT survey measures, how to calculate a CSAT score, when to send one, and what customer ratings can tell you.
A CSAT survey asks customers how satisfied they were with a specific interaction, transaction, product, or service. Its rating reflects respondents’ reported satisfaction, while an optional comment can help explain the score.
1. What is a CSAT survey?
A CSAT survey measures customer satisfaction with a particular interaction, transaction, product, or service. CSAT stands for “customer satisfaction,” and the survey usually asks a direct question such as, “How satisfied were you with your support experience?” (SurveyMonkey describes CSAT as feedback about a specific interaction or transaction.)
Customers might answer on a numbered scale, with labels such as “very dissatisfied” through “very satisfied,” or choose an emoji or written option. Many surveys add one optional comment question to learn why someone chose a rating. Keep the focus narrow: feedback about a recent support conversation is not the same as an assessment of a customer’s overall relationship with a company. CSAT is useful for checking individual touchpoints, but by itself it does not explain every factor behind satisfaction or loyalty.
2. How does a CSAT survey measure customer satisfaction?
A CSAT survey measures satisfaction by collecting customers’ own ratings of an experience and summarizing the positive responses. A common format uses a five-point scale from very dissatisfied to very satisfied, though organizations can use other scales or labels. (Qualtrics explains the five-point format and other survey channels.)
The wording should name the experience being rated. “How satisfied were you with the help you received today?” asks about support; “How satisfied are you with us?” is broader and may prompt different interpretations. Use a follow-up comment when you need context, but keep it optional if you want the rating to stay quick to answer. Because the metric comes from people who respond, it reflects respondents’ reported views, not necessarily every customer’s experience.
3. How is a CSAT score calculated?
A common CSAT score is the percentage of survey respondents who selected one of the defined positive ratings. On a five-point scale, teams often count ratings of 4 and 5 as positive, then divide that count by all valid responses and multiply by 100. (SurveyMonkey’s formula uses the top two categories as positive.)
CSAT = (positive responses ÷ total responses) × 100
For example, if 80 of 100 respondents choose a positive rating, the CSAT score is 80%. Neutral and negative responses stay in the denominator under this method, but only the top two ratings count as positive. Some teams report an average rating instead; that is a different calculation, so state which measure you use. Keep the scale, positive-response rule, and denominator consistent when comparing results over time.
4. When should you send a CSAT survey?
Send a CSAT survey soon after the experience you want customers to rate, but give them enough time to form an opinion. For a completed support interaction, ask shortly after it ends; for a product that needs to be used first, wait until the customer has had a reasonable chance to try it. (Qualtrics discusses timing for immediate interactions and longer-term product experiences.)
Other useful moments include after onboarding, a purchase, an appointment, or completion of a service. Choose the timing based on what you want to learn: a checkout survey can capture the purchase experience, while a later message can ask about product use. Avoid repeatedly surveying customers after routine interactions if that creates fatigue, and don’t ask so early that they have nothing meaningful to rate. (Zendesk notes that surveys should arrive while the experience is fresh, but not before customers can form an opinion.)
If you want to ask your own customers, Zigpoll is one option for ecommerce and Shopify brands, with on-site, post-purchase, email, and SMS surveys. As of November 15, 2026, its Lite plan includes 50 responses per month; paid plans start at $39 per month billed monthly or $29 per month billed annually (Zigpoll pricing). It may not suit teams that need a platform-neutral survey tool or a wider research program.
