Most CSAT surveys ask too many questions. Twenty questions, five pages, detailed breakdowns of every interaction. By question twelve the customer is clicking random numbers just to get through it.
You don't need twenty questions. You need the right ones.
CSAT measures satisfaction with a specific touchpoint. Ask customers to rate their experience 1 to 5, count the 4s and 5s, divide by total responses. Above 75% is good. Above 85% is excellent. The math is simple. The hard part is choosing what to ask.
Start here: the two essential questions
1. The satisfaction rating
"How satisfied are you with [your experience / this interaction]?" (1–5 scale)
This is your baseline. Every CSAT survey needs this question. Keep the scale consistent. Changing it mid-program breaks your ability to track trends.
2. The open-ended follow-up
"What's the main reason for your score?"
The number tells you what. This question tells you why. Without it, you know something's wrong but you can't prioritize which fix matters most. One rating plus one open-ended question is the minimum effective CSAT survey.
Go deeper when you need to
3. Resolution quality (post-support)
"How satisfied are you with how your issue was resolved?"
This separates the outcome from the agent. A customer can love the agent but still be frustrated the problem isn't fixed. That distinction drives very different actions.
4. Effort (post-support)
"How easy was it to get the help you needed?"
Customer effort is the strongest predictor of support-related churn. If people have to repeat themselves across channels or dig through help docs to find a phone number, they leave, even when the final answer was good.
5. Expectation match (post-purchase)
"Did the product meet your expectations?" (Exceeded / Met / Fell short)
When this score is low, the problem might not be your product. It might be how you described it. Expectation gaps are a marketing and positioning issue as much as a product one.
6. Value for price
"How would you rate the value for what you paid?"
Perceived value is independent of actual price. Someone who paid $500 and feels they got $1,000 worth is happier than someone who paid $50 and feels short-changed. This predicts upgrade and churn behavior better than the price tag alone.
Two questions most people overlook
7. Onboarding satisfaction
"How satisfied are you with how easy it was to get started?"
Onboarding friction is the silent killer of retention, especially in SaaS. Customers who struggle early churn disproportionately, even when the product is strong. Send this at the end of the onboarding flow, not weeks later when the memory has faded.
8. The one-change question
"If you could change one thing about your experience, what would it be?"
This is the most actionable open-ended question I've seen. Unlike "what could we do better?" (which invites wish lists), asking for the single most important change forces people to prioritize for you. Group the responses by theme and you have a roadmap straight from your customers.
Keep it short, send it fast
For a touchpoint survey: 3–5 questions, max. The two essentials plus 1–3 more based on context. Never exceed 10.
Timing matters. Post-support: send within hours while the experience is fresh. Post-purchase: wait 3–7 days so they've actually used the product. Post-onboarding: at the end of the flow. For broader relationship checks: quarterly.
The point
A good CSAT survey is short, well-timed, and always includes an open-ended question. Start with the satisfaction rating and "why?", then add more only if each additional question maps to a decision you'll actually make. Consistency beats sophistication. A simple 3-question survey run every quarter will teach you more than a 20-question survey run once a year.