
Net Promoter Score (NPS) is a customer loyalty metric calculated by subtracting the percentage of Detractors from the percentage of Promoters. It turns one standardized recommendation question into a score from -100 to +100.
Calculating NPS is simple. Getting a useful signal requires consistent timing, a representative response pool, and a process for acting on what customers tell you.
Customer happiness is a metric unlike any other. It is not founded only on revenue, bounce rate, or other traditionally mathematical ideas. Customer happiness is based on emotions, not data. While these emotions can be formulated as data, one of the first steps is to work out how.
Recognize the standard NPS email
Chances are, a company has asked you: How likely are you to recommend us to a friend or colleague? A true NPS survey presents that question on a scale from 0, not at all likely, to 10, extremely likely.
Chances are, you have received an email at some point that looks like a survey. The outcome of this question is used to formulate your Net Promoter Score, one of the metrics that measures the health of your company and the effectiveness of your customer success strategy. How do these responses translate to a solid representation of customer happiness?

The score gives customer success teams a consistent way to track recommendation intent. It does not prove customer happiness on its own, and it should not replace retention, expansion, support, or product-usage data. Treat it as one useful signal inside a broader set of customer success metrics.
Calculate NPS to determine customer happiness
Start by placing every response into one of three groups:
- Detractors score 0 to 6. They are less likely to recommend the company and may be at risk of leaving.
- Passives score 7 or 8. They are broadly satisfied but not enthusiastic enough to count as advocates.
- Promoters score 9 or 10. They are the customers most likely to recommend the company.
Detractors are customers unlikely to buy more, refer, and promote the business to those around them. If the customer base is filled with Detractors, the company may be in trouble. Passives are customers who are getting what they paid for and think the company offers a passable service. They may have enough gripes not to scream your praises from the rooftops, but they are not as dangerous as Detractors. Promoters are the customers you see retweeting product updates, commenting on blog posts, and publicly enthusing about their favorite companies. A company full of Promoters can create the effect of a great viral marketing campaign.

Use all valid responses as the denominator. Calculate the percentage of Promoters, calculate the percentage of Detractors, and subtract the second number from the first:
NPS = percentage of Promoters – percentage of Detractors
Passives are included in the total number of responses, but they do not appear directly in the subtraction. The result is a score, not a percentage.
A worked NPS calculation
Imagine you receive 493 valid responses: 173 Promoters, 286 Passives, and 34 Detractors.
- Promoters: 173 / 493 = 35%
- Detractors: 34 / 493 = 7%
- NPS: 35 – 7 = +28
The calculation is correct even though Passives do not appear in the final subtraction, because all 493 responses determine the two percentages.

What counts as a good NPS score?
There is no universal threshold that makes an NPS score good or bad. Industry, geography, customer segment, relationship stage, and survey method can all change the result. A score that leads one market may look ordinary in another.
Compare like with like. Track your own score over time with a consistent question, audience, and cadence. When you use an external benchmark, choose one that matches your industry and methodology. Also inspect response volume and sample composition before treating a small change as meaningful.
Survey NPS data from your customers
The collection method should fit the customer moment. A dedicated form works well for relationship surveys, an email can lower friction, and an in-product survey can capture a recent interaction. Whichever method you choose, keep the 0-10 question and timing consistent enough to compare one cohort with the next.
Use a dedicated survey form
A form provider can collect the score, an open-text reason, and the customer information needed for segmentation. Before choosing one, ask:
The quick way is to send a survey. You can use something as simple as SurveyMonkey or Google Forms to gather responses in a spreadsheet and work out your NPS. The longer way involves optimizing this process bit by bit. To make a good decision, answer questions about how many entries you expect, how the collected data is presented, what reports should be generated, where the form can be embedded, whether it looks professional, whether it can be branded or customized, and whether it integrates with the apps you use.
- How many responses do we expect?
- Can we export raw response data?
- Can we segment results by account, plan, lifecycle stage, or region?
- Does the form support a clean 0-10 scale and an optional follow-up question?
- Can it trigger the follow-up workflow we need?
- Can we keep the customer experience accessible, branded, and mobile-friendly?

Current survey platforms such as Typeform and SurveyMonkey support standard NPS questions. Google Forms can also work for a lightweight program when your team is prepared to calculate and segment the results separately.
Ask the first question in an email
An email survey can make the first response nearly effortless. Present the 0-10 choices in the message, then send the customer to a short follow-up form after the click. The follow-up should capture the reason for the score without turning a one-question pulse into a long questionnaire.
To reduce friction and make it more likely you will get a response, embed the survey directly into an email. When the recipient clicks a response, it can be directly logged without making them jump through hoops. You can set this up in an email marketing program, use Google Forms, or use Mailchimp to set up a poll or survey and gather data in an easily digestible way.

Use a delivery system that can associate the response with the right customer record, suppress duplicate invitations, and trigger an acknowledgment or follow-up. Mailchimp maintains current survey guidance, but the underlying selection criteria matter more than the vendor name.
Use an in-product or transactional survey
A transactional NPS survey follows a specific interaction, such as onboarding, a support case, a renewal, or use of an important feature. An in-product prompt can capture feedback while the experience is still fresh. Current platforms, including Intercom Surveys, support targeted survey delivery.
An in-app survey does not have to be an alternative to an email survey. You can use both when the programs are clearly labeled and the results stay separate. In-app surveys are especially useful for SaaS companies and mobile apps.

Do not mix transactional and relationship results without labeling them. A score after a support case answers a narrower question than a quarterly relationship survey. Keep the two series separate so an operational event does not distort the longer-term trend.
Time the survey consistently
Send transactional surveys soon after the event. For relationship NPS, choose a regular cadence that gives customers time to experience meaningful changes. Avoid surveying the same person so often that fatigue lowers response quality.
Make sure to send your NPS survey at a time when your product is fresh in your customer’s mind. This might mean an automated email triggered when a support ticket is closed, an account is upgraded, or soon after a service or transactional interaction. Finding the best time and highest response rate is specific to each company, so test what works for your customers.
Useful triggers include:
- after onboarding reaches a defined milestone;
- after a support case closes;
- before or after renewal;
- after a free trial or implementation period; and
- after a meaningful product or service interaction.
Once you have your survey, email, and software ready, decide whether to send the survey to a whole list or trigger it based on events. Key moments in the customer lifecycle include before or after subscription renewal, the end of a free trial, after the customer onboarding process, and after meaningful interactions with customer success or customer support.

Keep the survey focused
Lead with the standard recommendation question. Then ask one open-ended follow-up, such as: What is the primary reason for your score? That explanation is often more actionable than the number alone.
Test invitation wording, delivery channel, mobile layout, and timing. Keep the measurement method stable while testing one element at a time. Otherwise, you will not know whether the customer experience changed or the survey method changed.
Making changes to the way customers enter data and to the layout of the survey can improve response rates and give you more data to work with. This is worth A/B testing so you can learn whether customers prefer a numeric scale or another presentation, while keeping the standard NPS question and 0-to-10 scoring intact.

Turn NPS data into action
The score is the beginning of the process. Review open-text responses, group recurring themes, and segment the results by customer type or lifecycle stage. Look for the operational drivers behind the number rather than celebrating or defending the headline score.
Provided you have a way to collate the responses, it should be easy enough to calculate your NPS score. Follow the formula: the net percentage of Promoters, scored 9-10, minus Detractors, scored 0-6. Since scores fall inside a range of -100 to +100, work out category scores as percentages.
- Validate the response set. Check response volume, delivery failures, duplicates, and whether the sample represents the customer group you intended to survey.
- Segment the result. Compare meaningful cohorts without combining different survey types or customer moments.
- Review the reasons. Code open-text responses into themes such as onboarding, product reliability, support, value, or missing capabilities.
- Assign follow-up. Route urgent Detractor feedback to an owner, thank Promoters, and identify what could move Passives.
- Fix the recurring cause. Update the process, not only the individual customer record.
- Measure the next comparable cohort. Keep the question, audience, and timing consistent enough to see whether the change worked.
Take action to make improvements
Process Street is a single Compliance Operations Platform. Its Docs and Ops capability areas, together with built-in AI, let teams document the follow-up standard, route each response to the right owner, enforce deadlines, summarize recurring feedback, and keep an auditable record of the action taken. The point is not to collect a score. The point is to make the customer process improve because you collected it.
Once you know your NPS score, you know how urgent improvements may be. To know where you are going, you need to know where you are right now.
Frequently asked questions about calculating NPS
What is the formula for NPS?
Subtract the percentage of Detractors from the percentage of Promoters. Use all valid survey responses as the denominator for both percentages.
Is NPS calculated on a 0-10 or 1-10 scale?
Standard NPS uses a 0-10 recommendation scale. Scores from 0 to 6 are Detractors, 7 and 8 are Passives, and 9 and 10 are Promoters.
Do Passives count in the NPS calculation?
Passives count in the total number of valid responses, so they affect the percentages. They are not added to or subtracted from the final score directly.
How often should you send an NPS survey?
Use a regular relationship cadence or a clearly defined transactional trigger. The right frequency depends on how often customers have meaningful experiences with your product or service. Avoid sending surveys so often that customers become fatigued.
What should you do after calculating NPS?
Validate the sample, segment the score, review the reasons customers gave, assign follow-up, fix recurring causes, and compare the next equivalent cohort.