What is the level of customer satisfaction?
There is no universal level of customer satisfaction. A level is a score for a specific question, scale, group of customers, and time period. So “the level” means the score you measure for a defined experience, which you then judge against your own past results, the way the question was asked, and comparable studies. A single published figure from another company or industry is a weak basis for judging your own results.
What a satisfaction score means
Scale labels
Most satisfaction questions use a labeled scale. A common 5-point version looks like this:
| Score | Label |
|---|---|
| 1 | Very dissatisfied |
| 2 | Dissatisfied |
| 3 | Neutral |
| 4 | Satisfied |
| 5 | Very satisfied |
The labels define what a score means, so keep them the same across surveys. A score of 4 means “satisfied” on this scale, not “good” in some absolute sense.
CSAT share and average rating
Two common ways to summarize a level are:
- CSAT share. The percentage of responses in the satisfied range, usually the top two points on a 5-point scale. If 75 of 100 responses are 4 or 5, the CSAT is 75%.
- Average rating. The mean of all the scores, such as 3.8 out of 5.
Both are useful, but neither shows the full picture on its own. Two groups can share the same CSAT share yet differ in how many customers are neutral and how many are very dissatisfied. Look at the full distribution of answers alongside the summary. For how CSAT is calculated in more detail, see CSAT and KPI.
Why there is no universal benchmark
Satisfaction levels depend on many factors:
- Industry and product type. Expectations differ between, for example, a utility, a software product, and a restaurant.
- The question and the scale. A different wording or a 7-point scale produces different numbers.
- When and how you ask. A survey sent right after a support call usually reads differently from an annual survey.
- Who responds. A low response rate can make results look better or worse than the full customer base feels.
Some articles publish “good” satisfaction ranges. Treat those as rough guides only. A more useful comparison is your own trend over time, the same question across comparable customer groups, and published studies that used a similar method and question.
Read the level with context
Before you decide what a result means, check these points:
- Trend. Is the score rising, falling, or flat? A stable score at a low level may matter less than a sharp drop.
- Segments. Which groups are lower? A healthy overall score can hide a weak group, such as new customers or one region.
- Timing. Was the survey sent right after an interaction, or on a fixed schedule? The two answer different questions.
- Response. Who answered? If mostly the very happy or very unhappy customers responded, the level may not represent the typical customer.
- Reasons. What do low-scoring comments say? The reasons tell you more than the number alone.
What to do with a low or high level
If the level is low, look for the main drivers in the comments and in the segments with the lowest scores. Choose one problem to fix first, then measure again to see whether the score moves.
If the level is high, check that it reflects real performance rather than a survey that few dissatisfied customers completed. Keep measuring, so you notice if the score begins to slip.
In either case, avoid treating one number as a verdict. The level is a starting point for a question, not the answer to it.
Next steps
For a fuller guide to reading and reporting satisfaction results, see how to analyze customer satisfaction survey results and write the report. For the models that link satisfaction to its causes, see customer satisfaction models. For the broader set of survey types and metrics, see customer satisfaction surveys.