NPS / CSAT program audit

An NPS / CSAT program audit is a one-time expert review of how you measure customer satisfaction and loyalty (NPS, CSAT, CES and similar surveys): are the questions and scales formulated correctly (e.g. NPS is 0–10 with the proper split into promoters/passives/detractors, not "a five-point rating"); at what moment and after which event the survey is sent (in time, not right after purchase before real experience); to whom it is shown and whether the sample is biased (do only the happiest or angriest respond); what response rate you get and whether there are enough responses for conclusions; what happens to the responses — is there root-cause analysis, segmentation and "closing the loop" (acting on detractors), rather than just a number in a report; whether you compare the numbers correctly (against yourself over time, not against others' "industry averages" that are often incomparable). Honestly and importantly about the main point: this is an audit of the measurement itself, not a way to raise NPS, and not a guarantee of higher satisfaction. Crucially: improving the number in a report and improving real satisfaction are different things; a metric can be "gamed" with favorable timing or wording, but that is self-deception. NPS/CSAT reflect satisfaction but do not create it — it grows from the product, service and solving customers' problems, not from a survey. So our recommendations are about how to measure more honestly and usefully and what to do with the results, not "we will raise NPS by X points". Honestly about the method: on small samples and at a low response rate the numbers are unstable and easily misleading — we will point out where there is not enough data for conclusions. It is a snapshot in time: the program and tools change. And it is diagnosis, not building a survey system or working with the customers for you — implementation is separate. An important boundary: this is an audit of the satisfaction-measurement program, not a general UX or product audit (why customers are dissatisfied in essence — separate) — measurement shows that something is wrong but does not fix the cause. If you do not run surveys yet or have very few responses, there is nothing to audit and it is more useful to set up the collection first. Picture this: instead of "our NPS is 40, seems fine" you learn that the survey goes out an hour after purchase (before real experience), 3% respond and mostly happy ones, the scale is calculated incorrectly, and no one contacts the detractors — meaning the number cannot be trusted, and here is how to fix it. The base price starts from 12,000 ₽; it depends on the number of programs and survey channels.

• You measure NPS/CSAT but are not sure the numbers can be trusted.
• You suspect only the happy or only the angry respond.
• The survey arrives at the wrong time and few people respond.
• Responses are collected but there are no conclusions or action on detractors.

• Correctness of questions and scales (NPS 0–10, CSAT, CES)
• Timing and the trigger event of the survey
• To whom the survey is shown and sample bias
• Response rate and sufficiency of data for conclusions
• Root-cause analysis, segmentation, "closing the loop" with detractors
• Correctness of comparison (against yourself over time, not others' "averages")
• Where there is not enough data to trust the number
• A prioritized plan: how to measure more honestly and what to do with results

• A clear picture of whether your satisfaction numbers can be trusted
• Visibility of what distorts NPS/CSAT (timing, sample, scale)
• Priorities: what to fix in the measurement first
• Understanding where the problem is in the measurement and where in the product and service

• We clarify the survey programs, channels, goals and collect access to settings and results
• We check the questions, scales, timing, sample, response handling and comparisons
• We prepare a report with priorities and recommendations and review it with you

• Will NPS grow after the audit?

  That is not the goal of the audit, and honestly: raising the number in a report and raising real satisfaction are different things. A metric can be "gamed" with timing or wording, but that is self-deception. NPS/CSAT reflect satisfaction, they do not create it — it grows from the product, service and solving customers' problems. We make the measurement honest and useful, but "+X NPS points" cannot be promised.

• Can NPS be trusted with few responses?

  Often no. On small samples and at a low response rate the numbers jump around and easily mislead; sample bias (only the extremes respond) distorts the picture even more. We specifically show where there is not enough data for conclusions and where the number can be trusted.

• Will you fix low satisfaction and build the surveys?

  No. This is an audit of how you measure, not working with customers or building a survey system for you. The causes of dissatisfaction are the product and service (a separate UX/product review), and setting up surveys and implementation are done separately or by your team.