Net Promoter Score (NPS) is an easy way to measure customer relationships by using one question: "How likely is it that you would recommend our company/product/service to a friend or colleague?"
Scoring is typically on a scale of 0 (worst) to 10 (best).
We recommend Net Promoter Score as our #1 metric for rollup up results, and this page explains why.
NPS has been widely adopted and is in use by more than two thirds of Fortune 1000 companies.
NPS aims to measure the loyalty that exists between a producer and a consumer. The producer can be a company, employer, interviewer, etc. The consumer can be a customer, employee, interviewee, etc.
An NPS can be as low as −100 (every respondent is a "detractor") or as high as +100 (every respondent is a "promoter"). A positive NPS is generally good. A NPS of +50 is generally excellent.
The calculation is here.
Net Promoter Score is related to areas and questions such as:
-
Customer satisfaction ratings, such as "How is this, on a scale of 1-5 stars?"
-
Approval categorization bins, such as "Do you want this? Yes / No / Maybe".
-
Scorecard grades, such as "What's your grade for this? Grade A is best, F is worst".
-
Loyalty sentiment, such as "How committed are you to this? Very / Somewhat / Not".
-
Outreach potential, such as "What percentage of your peers would like this?".
-
NPS is a widely-known high-level good-enough metric, in active use by thousands of companies, agencies, industries.
-
NPS is fast and easy use and analyze, because it proxies for satisfaction, approval, scorecards, retention, loyalty, growth, and more - see below for examples.
-
NPS is the most-capable for spanning many contexts. We're able to use it for simple quick user interviews using mockups and sketches, and we're also able to use it at very large scale for ongoing sophisticated research involving user engagement involving many projects and many timelines.
-
NPS relies on user sentiment and user prediction of future actions, rather than actual behavior.
-
NPS is statistically imperfect/overgeneralized/vague and companies should be asking more-exact more-specific more-targeted questions that result in better data.
-
NPS takes too much time, because the user needs to have an experience, then agree to respond.
-
NPS relies on user sentiment and user prediction of future actions, rather than actual behavior.
-
We're fine with this because we use NPS just as a first approximation rollup, and we help adjust for bias by also using real data from real actions.
-
We can also run analytics that compare a user's prediction and actual behavior; this is an easy valuable way to improve our understanding of users' attitudes compared to actions, and to surface areas where there are gaps.
-
-
NPS is statistically imperfect/overgeneralized/vague and companies should be asking more-exact more-specific more-targeted questions that result in better data.
-
We're fine with this because we highly value having a one top-level well-known metric that works across all areas, versus having multiple top-level metrics, or lesser-known top-level metrics, or disparate top-level metrics.
-
We can adjust for this by rolling down to more-specific statistics and more-targeted questions, and we use a range of tactics such as involving subject matter experts, doing customer discovery interviews, and providing free form reponse capabilties; this enables us to use NPS at the top, and expand as needed to lower-level specifics.
-
-
NPS takes too much time, because the user needs to have an experience, then agree to respond.
-
We use NPS in the traditional way and also use NPS in a rapid way because we specialize in agile/lean development and value fast decision guidance.
-
We can show a user a few one-page mockups, and ask the person for NPS for each one; this takes just a couple minutes per mockup. The fast answers are valuable for us because we emphasize improvement and iteration; we care less about the number score, and more about the approximate trendline of whether we're increasing or decreasing. We then improve the work by creating high-fidelity prototypes, then real launches, and at each step we can ask for NPS to help firm up the score so it becomes more accurate and more experiential.
-
We especially like the metrics and descriptions created by Andreessen Horowitz, a leading Silicon Valley investment firm for startups. These metrics can be good for a wide range of companies and projects of all sizes.
