CSAT

What is CSAT?

CSAT (Customer Satisfaction Score) refers to a metric that measures customers' satisfaction with a particular service or a particular contact. It is collected with a single question immediately after the event, usually on a scale of 1 to 5. The question asked is, for example, how satisfied the customer was with the resolution of their request. The figure reported is generally the share of satisfied answers among all answers.

Typical occasions are a customer service request, the introduction of software, a delivery or a consulting appointment. The metric belongs in the metrics report of customer service and product management and occasionally appears in the investor update. In a company sale, it is among the data that an acquirer requests in commercial due diligence, the review of market, customers and competition. In service contracts it is sometimes set as a quality target in a service level agreement (SLA).

Calculating CSAT: formula and scales

Usually the answers in the top two levels of the scale are counted (top-2 box).

Formula: CSAT = (number of answers of 4 or 5 ÷ number of all answers) × 100, stated as a percentage

Alternatively the mean of all answers is stated, for example 3.8 out of 5. Both forms are in use but cannot be converted into each other; which one is used is part of the definition of the metric. Besides the five-point scale, seven-point and ten-point scales as well as symbol scales are common. Within a time series the scale must not change.

Every CSAT comes with the response rate and the number of answers. With a low response rate, it is often mainly customers with very good or very bad experiences who answer, and the figure reflects the whole customer base only to a limited extent. With a few dozen answers, the figure fluctuates so strongly by chance alone that changes of a few percentage points say nothing.

CSAT survey: collection and legal framework

CSAT measures a moment. The question is therefore asked as soon as possible after the contact, in the product, by email or at the end of a conversation; the longer the interval, the more the survey measures memory. All contacts of a type or a random selection are surveyed, not only successfully closed cases, because otherwise precisely the dissatisfied customers drop out of the measurement.

In Germany the law sets limits on collection. According to case law, a satisfaction survey by email can count as advertising, which may be inadmissible without the recipient's consent; whether an exception for existing customers applies depends on the circumstances. If answers are assigned to individual persons, the requirements of the EU General Data Protection Regulation (GDPR; in German Datenschutz-Grundverordnung, DSGVO) on legal basis, information and storage period apply. Surveys in the product or during an ongoing contact do not raise the question of advertising by email, but the data protection requirements apply there too.

Informative value and limits

As an absolute figure, CSAT is comparable only to a limited extent, because wording, scale, timing and channel strongly influence it. It becomes meaningful as a time series under constant conditions and in comparison between segments, for example between new customers and existing customers or between product lines. The negative answers are particularly valuable: every rating of 1 or 2 is a reason to get back to the customer. The free-text comments show causes that the metric itself cannot reveal.

In business with corporate customers, it is mostly users who answer, while other people decide on contract renewal. A high CSAT in customer service therefore does not rule out a cancellation at decision-maker level.

Significance in due diligence and valuation

For company valuation, CSAT is not a direct input. Investors and acquirers assess customer retention and revenue development in the existing base, measured by the cancellation rate (churn rate) and the retention rate (retention rate). CSAT is a leading indicator of these figures and serves as evidence when a company attributes an improvement in customer retention to a specific measure.

In commercial due diligence, satisfaction data is reviewed in context: methodology and response rate, development over several quarters, values of the largest customers and reconciliation with cancellations and renewals. The acquirer often supplements the data with its own conversations with selected customers. If these deviate markedly from the reported values, the credibility of the entire metrics report suffers.

Worked example: calculating CSAT in customer service

A software provider handles 400 customer service requests in a month. After every request, whether resolved or not, the person making the request receives the rating question. 160 people answer: 64 with 5, 40 with 4, 24 with 3, 18 with 2 and 14 with 1.

• Response rate: 160 ÷ 400 = 40.0%
• Satisfied answers: 64 + 40 = 104
• CSAT: 104 ÷ 160 × 100 = 65.0%
• Negative answers: 18 + 14 = 32, that is, 32 ÷ 160 = 20.0%
• Mean: (5 × 64 + 4 × 40 + 3 × 24 + 2 × 18 + 1 × 14) ÷ 160 = 602 ÷ 160 = 3.76
• Statistical margin of error at 95% confidence (confidence interval), approximately: 1.96 × square root of (0.65 × 0.35 ÷ 160) = 1.96 × 0.0377 = 0.074, that is, about 7.4 percentage points in either direction

The margin of error indicates how far the measured value can deviate from the value for all customers purely because these particular people happened to answer; distortions from one-sided response come on top. If CSAT rises to 69% the following month with a similar number of answers, the change lies within this margin of error and does not in itself prove an improvement. More informative are a trend over several months and the 32 negative answers: if they mostly concern the same topic, for example errors in data migration, the cause has been found that the overall figure conceals.

Distinction: CSAT, NPS, KPI and Customer Effort Score

• NPS: The Net Promoter Score asks on a scale of 0 to 10 about the willingness to recommend. It results from the share of promoters (9 and 10) minus the share of detractors (0 to 6). It is stated as a number between minus 100 and plus 100, not as a percentage.
• KPI: a metric with a target value, responsibility and steering effect. CSAT becomes a KPI when a company, for example, measures its customer service against a target value.
• North star metric: measures what customers actually do in the product, whereas CSAT measures how they rate a single contact. A survey is therefore rarely suitable as a guiding metric.
• Customer Effort Score: measures how much effort a customer had to make for a request and complements CSAT where simplicity matters more than perceived friendliness.
• CSAT as an exam: In South Korea, CSAT also denotes the nationwide university entrance exam (College Scholastic Ability Test). It shares only the abbreviation with the metric.

Common mistakes

• Survey only for resolved cases. If unresolved or abandoned requests drop out of the survey, the most dissatisfied group is missing, and the figure rises without any improvement.
• No response rate reported. Without the response rate and the number of answers, a CSAT cannot be interpreted.
• Bonuses without a secured selection. If bonuses depend on CSAT while employees themselves decide who is surveyed, the sample becomes distorted.
• Change of scale or question. Any change to the wording or scale breaks the time series.

Frequently asked questions

What is a good CSAT value?

As a rough rule of thumb, a value from about 75% satisfied answers (top-2 box on a five-point scale) counts as good. This mark is not universally valid, because industry, channel and question strongly influence the figure. A comparison with the values of other companies (benchmarking) also presupposes the same question, scale and survey timing. The benchmark is one's own time series under the same conditions; the share of negative answers deserves more attention than the overall figure.

How is a CSAT survey structured?

Of a rating question and an optional free-text field. The question refers to the specific contact, for example the resolution of a service request, and is answered on a scale from 1 (very dissatisfied) to 5 (very satisfied). The free-text field asks for a brief explanation. Additional questions generally lower the response rate and turn the contact survey into a general customer study.

What is the difference between CSAT and NPS?

CSAT measures satisfaction with a single contact, NPS the willingness to recommend and thus the overall relationship. One metric can therefore not replace the other. A combination of ongoing CSAT after contacts and an NPS survey at longer intervals is usual.

Is CSAT data expected in the data room?

In a company sale, often, when customer retention is part of the equity story. Methodology, response rate and time series then belong in the data room, not just the current value.

The Roemer Capital view

Roemer Capital uses satisfaction scores in investor documents mainly as evidence, rarely as a statement in their own right. If a client attributes a falling cancellation rate to an improved onboarding, Roemer Capital shows the course of CSAT after onboarding next to the development of cancellations. Only both together turn a claim into a comprehensible connection.

Before a company sale, Roemer Capital checks with the management whether the satisfaction data will stand up to commercial due diligence. This includes whether the survey covers all contacts, how high the response rate is and whether the values fit cancellations and contract renewals. Gaps in the methodology can be closed before the process, but in due diligence they can only be explained.