CSAT stands for Customer Satisfaction Score, a survey metric measuring how satisfied a customer was with a specific interaction. You calculate a csat score by dividing satisfied responses by total responses and expressing the result as a percentage. It is a fast, transactional instrument for GTM teams checking whether a touchpoint delivered value.
Key Takeaways
- CSAT measures satisfaction with a specific interaction, not loyalty or long-term sentiment.
- The csat formula is satisfied responses divided by total responses, shown as a percentage.
- Scale choice changes the number, so cross-company comparison is unsafe without context.
- Trigger csat surveys at transactional moments: onboarding, support, purchase, launch.
- Segment CSAT by plan, cohort, channel, and ticket type so the average does not hide problems.
- CSAT answers "did this touchpoint work"; route signal into product and marketing action.
What Does CSAT Stand for and What Does It Measure?
CSAT stands for Customer Satisfaction Score. It is a metric that captures how satisfied a customer was with a discrete experience: a support reply, an onboarding flow, a purchase, or a new feature. Unlike relationship metrics that ask how a customer feels about your company overall, CSAT is scoped to one moment. For a startup growth or GTM team, that scoping is the point. It tells you whether the last interaction you engineered actually landed, while the memory is fresh and the signal is still actionable.
What Is the CSAT Formula?
The csat formula is simple. You count the responses that fall on the "satisfied" side of your scale and divide by the total number of responses, then express the result as a percentage. In plain terms: satisfied responses divided by total responses, times 100. A csat score of 80 percent means eight in ten respondents were satisfied with the interaction you measured.
How Do You Calculate CSAT with a Worked Example?
Here is how to calculate csat on a 5-point scale where responses of 4 and 5 count as satisfied. Suppose a post-support csat survey returns 200 total responses: 110 say 5, 50 say 4, 20 say 3, 15 say 2, and 5 say 1. Satisfied responses are 110 plus 50, or 160. The math is 160 divided by 200, which equals 0.80, or an 80 percent csat score. The same responses on a binary scale, where only a "yes" counts, could look very different, which is why the scale matters more than the raw percentage.
What Scale Choices Change the Number?
The scale you pick changes the number more than the underlying sentiment does. A 5-point scale (1 to 5) is common for support. A 1-10 scale lets you compute a mean and a top-box rate. A binary thumbs up or down is the cleanest for transactional prompts but loses nuance. Because "satisfied" is defined differently on each scale, two companies can report a "good" csat score while measuring different things. Cross-company comparison is unsafe unless you also compare scale, wording, and timing.
Where in the Lifecycle Should You Trigger a CSAT Survey?
CSAT works best as a transactional survey fired right after a specific event. The four high-value triggers for most GTM teams are post-onboarding, post-support-ticket, post-purchase, and post-feature-launch. Each captures a different slice of the journey. Post-onboarding CSAT tells you if activation worked. Post-ticket CSAT tells you if support resolved the issue. Post-purchase CSAT tells you if the buy experience held up. Post-launch CSAT tells you if the new capability landed.
How Does Transactional Timing Differ from a Relationship Survey?
A transactional csat survey fires immediately after one event and asks about that event only. A relationship survey asks about the customer's overall feeling toward your company and fires on a calendar cadence, such as quarterly. Transactional CSAT is high-signal and low-effort because the experience is recent. Relationship surveys are broader but noisier and easier to ignore. For a startup, start with transactional CSAT; add relationship measurement only once you have volume and a reason to track trend, not just events.
How Do You Write the CSAT Question and Follow-Up?
Keep the question specific and the scale explicit. A clean prompt is "How satisfied were you with the support you just received?" with your chosen scale underneath. The mandatory free-text follow-up is "What is the main reason for your score?" This open field is where the real signal lives, because the number tells you there is a problem and the text tells you which one. Do not skip it. Without the follow-up, a dropping csat score is a mystery you cannot act on.
What Are the Sample Size and Response-Rate Realities for Small Startups?
Small startups rarely get the response rates enterprise benchmarks assume. A 5 to 15 percent response rate on a transactional prompt is a realistic range for an early team, not a failure. That means a 200-response month may be a strong quarter. The practical rule is to watch trend and segment rather than chase a statistically perfect sample. A csat survey with 30 responses from a high-value cohort can be more useful than 2,000 mixed responses you cannot segment.
How Should You Segment CSAT So the Average Does Not Hide the Problem?
- By plan: enterprise and free users experience different product surfaces.
- By cohort: signup month or acquisition channel predicts different expectations.
- By support channel: chat, email, and phone produce different satisfaction patterns.
- By ticket type: billing issues and bugs are different satisfaction problems.
A single company-wide csat score averages these apart. The average looks fine while one segment is bleeding. Segment before you report, and report the worst segment first so it gets fixed.
How Does CSAT Differ from NPS and a Churn Survey?
CSAT asks about satisfaction with a specific interaction. NPS asks about likelihood to recommend and is a relationship metric. A churn survey asks why a customer left and fires on cancellation. They answer different decisions: CSAT tells you if a touchpoint worked, NPS tells you about growth potential through referrals, and a churn survey tells you why revenue walked. Use csat vs nps as complementary, not competing, signals. The table below keeps the distinctions crisp.
| Metric | What it asks | When it fires | Decision it informs | Main limitation |
|---|---|---|---|---|
| CSAT | How satisfied were you with this interaction? | Right after a specific event | Did the touchpoint deliver value? | Scoped to one moment; scale choice skews comparison |
| NPS | How likely are you to recommend us? | Periodic relationship cadence | Referral and growth potential | Correlates loosely with revenue; vague driver |
| Churn survey | Why did you cancel or downgrade? | On cancellation or downgrade | Why revenue is leaving | Only captures leavers; survivorship bias on stayers |
Which Question Answers Which Decision?
Match the instrument to the decision instead of stacking surveys. If the decision is "should we rewrite the onboarding flow," use post-onboarding CSAT plus its free-text. If the decision is "should we invest in a referral program," use NPS trend. If the decision is "why did pipeline accounts churn," use the churn survey. A common mistake is asking an NPS-style question after support and calling it CSAT. The question wording determines which decision the data can honestly answer.
How Do You Route CSAT Signal into Product and Marketing Action?
- Close the loop with low scorers by triggering a human follow-up within one business day.
- Route promoter responses into review generation by asking happy users for a public review.
- Build save plays for low scorers, such as a billing credit or a guided setup call.
- Feed free-text themes into the product backlog ranked by frequency and segment impact.
- Report segment CSAT to GTM leads weekly so trends are caught before they compound.
The loop only works if someone owns the follow-up. A csat survey that collects data and routes nowhere is vanity reporting with extra steps.
How Do You Instrument CSAT in Your Stack?
Instrumentation is where GTM engineering earns its keep. Set event triggers in your product analytics or lifecycle tool so the csat survey fires on the right action. Tag each response with the plan, cohort, channel, and ticket type so segmentation is free later. Pipe responses to your warehouse or CRM so CSAT lives next to revenue and support data. Without tagging and piping, you end up exporting CSVs by hand and the signal goes stale before anyone reads it.
What Are the CSAT Failure Modes to Avoid?
- Survey fatigue: too many prompts train users to ignore or dismiss them.
- Gaming by support reps: reps who only close easy tickets inflate the score.
- Vanity reporting: a single rising average that hides a suffering segment.
- Surveying only happy users: triggering only on success sessions biases the number up.
Each failure mode makes the csat score look better than the product feels. Guard against them with random sampling, segment transparency, and a leader who reads the worst segment first.
Frequently Asked Questions
What Is a Good CSAT Score?
A good CSAT score depends entirely on your scale and audience, so avoid borrowing someone else's benchmark. On a 5-point scale where 4 and 5 count as satisfied, many teams treat 75 to 85 percent as a reasonable working target, but that is a rule of thumb, not a law. The more useful question is whether your csat score is trending up within a specific segment. Compare your own cohorts over time before comparing to any external number you did not measure yourself.
How Is CSAT Different from NPS?
CSAT measures satisfaction with a single interaction, while NPS measures likelihood to recommend your company overall. CSAT is transactional and fires right after an event; NPS is relational and fires on a cadence. CSAT answers "did this touchpoint work," and NPS answers "would this customer help us grow." The csat vs nps distinction matters because they catch different problems. A customer can be satisfied with support yet unlikely to recommend if pricing or the core product disappoints.
How Do You Calculate CSAT from Raw Responses?
You calculate CSAT by counting satisfied responses, dividing by total responses, and multiplying by 100. On a 5-point scale, define satisfied as 4 and 5, then sum those counts. If 160 of 200 respondents chose 4 or 5, the csat score is 160 divided by 200, or 80 percent. Always state your satisfied threshold when you report, because the csat formula is meaningless without it. The same raw data can yield different scores under different scale definitions.
When Should a Startup Send a CSAT Survey?
A startup should send a csat survey immediately after a specific, high-stakes interaction rather than on a vague schedule. The four strongest moments are post-onboarding, post-support-ticket, post-purchase, and post-feature-launch. Transactional timing captures fresh, accurate sentiment and keeps the survey short. Avoid quarterly relationship blasts until you have enough volume to see trend. Early on, one well-placed csat survey after support will teach you more than a company-wide send.
Why Include a Free-Text Follow-Up on a CSAT Survey?
The free-text follow-up turns a number into a cause. The csat score tells you satisfaction dropped, but the open question tells you whether it was billing confusion, a slow reply, or a missing feature. Without it, low scores are unactionable and teams guess. Keep the prompt simple, such as "What is the main reason for your score?" Then route those themes into the product and support backlog. The free-text field is the cheapest user research a startup can run.