What Is an NPS Survey? Implementation Steps, Question Examples, Tabulation, and Use
Published:
Last Updated:
Category: CRM, LTV & Customer Management
Published:
Last Updated:
Category: CRM, LTV & Customer Management
Authors: Shusaku Yosa
How much do customers trust your company and want to recommend it to others? An NPS survey lets you measure exactly this with a single number. The questions are simple, but merely running it tends to end up as "a score comes out and that's it." In this article, we organize what an NPS survey is, then explain the implementation steps from objective setting through tabulation and use, ready-to-use question examples, how to tabulate the score, and how to use the results to drive improvement, all in line with practice.
An NPS survey is a survey that measures customer loyalty (attachment to and trust in a company or brand) using a metric called NPS (Net Promoter Score). NPS is also translated as "customer recommendation level" and quantifies the degree to which customers want to recommend a product or service to friends or colleagues.
Its greatest feature is that the survey is extremely simple. The core question is just one, and the answer is simply choosing a number from 0 to 10. As a result, the burden on respondents is small, the response rate is high, and it is easy to measure continuously. Also, because it measures with a common yardstick of 0 to 10, it is easy to benchmark against industry averages and competitors, which is another reason many companies adopt it.
A similar survey is the customer satisfaction (CS) survey, but the two measure different things. A CS survey measures current satisfaction, that is, "are you satisfied now," whereas NPS measures loyalty that connects to future behavior, that is, "do you want to recommend it to others." NPS is said to correlate more strongly with business performance such as repeat purchases and continued use, making it easier to use as a management metric.
An NPS survey generally proceeds in the following five steps. More than producing the score itself, the design and use before and after it govern the outcome.
First, clarify "what you are surveying for." Whether you want to know the recommendation level for the whole product/service, or the evaluation of a specific touchpoint (such as after purchase or after support), changes both the survey type and the questions. Together with this, decide whom to ask (all customers, repeat customers, a specific attribute, and so on). How you choose the target is an important decision that directly affects how you later use the numbers.
NPS surveys come in two types: "relational surveys" and "transactional surveys." Use them according to your objective, and ideally combine both.
An effective flow is to first grasp the overall picture and issues with a relational survey, then dig into individual touchpoints with a transactional survey.
In addition to the basic question asking the recommendation level, combine a free-response question asking the reason and attribute questions used for analysis. Because too many questions lower the response rate, the trick is to narrow it down to a guideline of seven questions or fewer overall. Specific question examples are described below.
Deliver via the method easiest for target customers to answer, such as email, SMS, web forms, or in-app display. Because the reliability of the score depends on the number of responses, set the number of requests higher on the premise that responses are hard to gather, and also consider countermeasures for cases where they do not gather.
Tabulate the responses to calculate the NPS score, and analyze it together with the free responses and attributes. What matters is not ending with "measuring and done." Sharing the results with the front line and translating them into specific improvement actions is the purpose of the survey.
NPS survey questions are broadly composed of three types: a "recommendation question," a "free response asking the reason," and "attribute questions."
This is the core question of NPS, also called "the ultimate question." Always include it.
Question example: "How likely are you to recommend this product (service) to a close friend or colleague?" (Answered on an 11-point scale from 0 = not at all likely to 10 = extremely likely)
The score alone does not tell you "why that number." Adding a free response asking the reason yields hints for improvement.
Question example: "If you don't mind, please tell us the reason for the score you gave."
Because free responses tend to burden respondents, narrow the number of such questions and place them in the latter half of the questionnaire.
To increase the angles of analysis, add questions asking attributes (age group, length of use, and so on) and satisfaction at each touchpoint. For example, for SaaS, having respondents rate "features," "usability," "price," and "support" individually makes strengths and weaknesses concretely visible.
Once responses are gathered, classify respondents into three groups according to their scores.
Once classified, calculate the NPS score with the following formula. Passives are not used in the calculation.
NPS = percentage of promoters (%) - percentage of detractors (%)
For example, if out of 100 respondents there are 60 promoters, 30 passives, and 10 detractors, the NPS is "60% - 10% = 50." The score ranges from a maximum of +100 to a minimum of -100, and it can even go negative when there are many detractors. In Japan there is a tendency to choose the middle value, so detractors tend to come out higher; therefore the basic approach is to view it by trend and comparison rather than absolute value.
NPS is not something to judge by a single number, as in "50, so it's good or bad." There is meaning in measuring continuously with the same question and seeing whether it is trending up or down and how it changed before and after a measure. Furthermore, by comparing against the industry average and competitor scores (benchmarks), you can grasp your company's standing. Because average values vary greatly by industry, the iron rule is to make comparisons within the same industry.
Combining the quantitative data of the score with the qualitative data of free responses reveals "why that score." Dissatisfactions common in detractors' free responses are issues to improve with priority. Conversely, points that promoters rate highly are your company's strengths to extend.
Rather than keeping the results within management alone, it is important to share them promptly with front-line staff who interact directly with customers. By conveying not just the score number but also the specific reasons as a set, it becomes clear which behaviors to continue and which to fix, and autonomous improvement arises. Positioning NPS as a key metric (KPI) and building a mechanism to run the measure-share-improve cycle leads to results.
An NPS survey quantifies customer loyalty with a simple single question and can be used for correlation with business performance and for benchmark comparison. The implementation steps are five: setting the objective and target, choosing the survey type, designing the questionnaire, running it, and tabulating and using the results. The basics are to find the score by subtracting the percentage of detractors from the percentage of promoters, and to read it by trend and comparison and by quantitative x qualitative.
What matters most is not producing the score itself, but sharing the results with the front line and connecting them to improvement. First narrow the objective to one, start small with simple questions, and run the cycle of measurement and improvement.
A clear explanation of how CPM advertising (impression-based billing) works and its calculation formula. Covers the diff...
A clear explanation of what O2O (Online to Offline) means and how to read it. Covers the differences from OMO, omnichann...
Organizes the relationship between ROI and cost-effectiveness and explains how to calculate ROI with examples. Covers th...
