How to Choose a Survey Scale

Choose a survey scale after defining the decision. A familiar scale is helpful, but consistency and a clear question matter more than chasing a universal “best” format.

Start with the experience

Name what the person is evaluating: recommendation, satisfaction with an interaction, or effort on a task. The metric should fit the experience, not the other way around.

Use the metric chooser when the decision is still unclear.

Common choices

NPS: a 0-to-10 recommendation question for the broader relationship.
CSAT: a satisfaction score after a specific experience, often on a 1-to-5 scale.
CES: an effort question tied to a task, often on a 1-to-5 or 1-to-7 scale.
Agreement scales: useful for a clear statement, but easy to interpret as a general attitude if the context is missing.

The NPS, CSAT, and CES comparison explains where each metric fits.

Explain the endpoints

Label both ends in plain language. “1 = very difficult” and “5 = very easy” gives the respondent a frame; a row of numbers alone does not.

If you use a written scale, keep the direction stable. Switching from “higher is better” to “higher is harder” makes the answer harder to read.

Keep comparisons honest

Keep the question, scale, audience, channel, and send moment stable when comparing rounds. A change in any one of them can move the score without a real experience change.

Record the threshold when you calculate a percentage, such as which CSAT answers count as satisfied.

Add context without adding noise

A score shows direction but not the reason. Add one optional open question, such as “What was the main reason for your answer?”. Avoid combining several experiences into one rating.