How to design a CSAT survey

Start with the moment

A CSAT survey measures satisfaction with something specific. Before writing a question, choose the moment: a store visit, a delivery, a support call, a meal, a service appointment. A survey that asks about "your experience with us" in general produces a score nobody can act on.

Write the question

• "How satisfied were you with your visit to [store] today?"
• "How satisfied were you with your delivery?"
• "How satisfied were you with how we handled your request?"

Name the moment, use plain words, and avoid adjectives that lead the answer such as "our friendly team".

Choose the scale

A 5-point scale from "very dissatisfied" to "very satisfied" is the most common choice and reads well on a phone. Label every point or at least the ends and the middle. Stars or faces work for quick in-store screens, but keep the same five levels so the data stays comparable.

Add the follow-up

Follow the score with one open question. "What is the main reason for your rating?" works for everyone. You can tailor it: "What went wrong?" for 1 and 2, "What could we have done better?" for 3, "What did you like most?" for 4 and 5.

Diagnostic questions, sparingly

If you need structure, add one or two short questions about the main drivers of the moment, such as wait time, staff helpfulness or product availability. Only add a question if someone will use the answer to make a decision.

Capture context automatically

Pass store ID, date, channel and transaction type into the survey as hidden fields rather than asking customers. It shortens the survey and makes store-level reporting reliable.

Channel and timing

Send soon after the moment: within hours for a visit or delivery, at the end of a support contact for service. Use the channel the customer already uses with you, and make the first question answerable in one tap from the message itself where possible.

A sample layout

1. How satisfied were you with your visit to [store] today? (1 very dissatisfied to 5 very satisfied)
2. What is the main reason for your rating? (open, optional)
3. How long did you wait at the billing counter? (under 5 minutes, 5 to 10, over 10)
4. May we contact you about your answer? (yes / no)

Calculate it consistently

The common method is the share of 4 and 5 answers out of all answers. Some teams use the average instead. Either is fine; write the method down and show the number of responses beside every score.

Test before launch

Send the draft to staff and a few customers. Check it on a cheap phone, time it, and read the answers to see whether the questions were understood. Fix confusing wording before the first real send, because changing it later breaks the trend.

Common mistakes

• Asking about the whole relationship in a CSAT survey.
• Ten questions where three would do.
• Unlabelled scales.
• Asking customers for details you already have.
• Changing wording after launch.

Common questions

Checklist

• One moment chosen.
• Question names the moment.
• Labelled scale.
• Open follow-up.
• Context in hidden fields.
• Method documented.
• Tested on a phone.

In short

Design a CSAT survey around one moment, ask one clear satisfaction question on a labelled scale, add an open follow-up and few diagnostics, pass context automatically and keep the method fixed.

Doing this with Falcon

Falcon does not run surveys or feedback forms. It analyses the unprompted feedback customers already leave in Google Maps reviews of your locations and Google Play reviews of your app: each review is classified into themes, sub-themes and specific issues from your own taxonomy, with the sentiment of each mention and the exact quote behind it. A CSAT survey hears from customers you invite; reviews add those you did not. Falcon's themes and negative share by store let you check whether low CSAT stores also stand out in reviews, and for which topics.