A Single Score Won’t Tell You What Customers Think

A customer score tells you something happened. Only customers can tell you why.

The owner of a small software company proudly shares the latest customer survey result at the monthly team meeting: a Net Promoter Score of 62, up from 54 last quarter. Everyone claps. Someone asks what changed. Nobody is quite sure. Someone else asks what the team should do differently next quarter. Again, silence.

The Net Promoter Score, or NPS, is one of the most popular customer metrics in the world. It comes from a single question: “How likely are you to recommend us to a friend or colleague?”, answered on a scale from 0 to 10. Its appeal is obvious. It’s quick for customers, simple to calculate and easy to share in a meeting.

But as the software company found, a number on its own rarely tells you why customers feel the way they do, or what to change. And research has questioned some of the bigger claims made about what that number can predict.

A customer score tells you something happened. Only customers can tell you why.

What the research found

The NPS was introduced in 2003 by the loyalty consultant Fred Reichheld in a Harvard Business Review article titled “The One Number You Need to Grow”. He argued that this single question was a better predictor of company growth than other customer measures, such as satisfaction.

In 2007, a team led by Timothy Keiningham, with academics Bruce Cooil, Tor Wallin Andreassen and Lerzan Aksoy, tested that claim in the Journal of Marketing. Using longitudinal data from 21 Norwegian firms and more than 15,500 customer interviews, and revisiting industries Reichheld had used as examples, they could not replicate his assertions regarding the “clear superiority” of NPS compared with other measures. The paper later won a prestigious award from the Marketing Science Institute.

Later analyses have reached a more balanced view. NPS tends to correlate closely with customer satisfaction, and both measures predict business outcomes to a broadly similar degree. Sometimes NPS performs best; sometimes it doesn’t. It isn’t useless. It simply isn’t the single magic number it was first presented as.

It’s worth noting that Keiningham worked for the research firm Ipsos at the time, which offered other loyalty measures, and that NPS supporters have defended the metric’s simplicity and practical value. The debate continues. But for a small business, the more useful lesson is less about which number is best and more about what any single number can and can’t tell you.

One number can tell you something changed, but not why.

From score to story

The most valuable part of any customer survey is usually the part that gets the least attention: the open question that follows the score. “What’s the main reason for your answer?” turns a number into something you can act on.

In practice, that means a few habits.

Read every comment, especially the middling ones. Customers who score you 7 or 8 often give the most useful feedback. They like you, but something is holding them back from enthusiasm. Their comments frequently point to small, fixable frustrations.

Group comments into themes. Once a month, sort the reasons into a handful of categories, such as speed, communication, price or quality. A rising theme is a far clearer signal than a score moving by a few points.

Close the loop. Contact some customers who gave low scores, thank them, and ask a follow-up question. Many will be surprised and pleased that anyone read their answer. Some will tell you exactly what would win them back.

Watch behaviour as well as opinions. Repeat purchases, referrals, cancellations and complaints tell you what customers actually do, which, as we explored in Why Design Thinking Works, When It Works, doesn’t always match what they say.

INFOGRAPHIC GOES HERE

From number to action. ). Caption: Keiningham et al. (2007) found no evidence that NPS was clearly superior to other customer measures in predicting growth.

A score → A reason → An action.

Using customer scores well

If you already use NPS or a satisfaction score, there’s no need to abandon it. Simple, consistent measures are useful for spotting broad trends over time. But treat the number as a smoke alarm rather than a diagnosis. When it moves, the real work is finding out why.

If you don’t use any customer measure yet, start small. A single question with a follow-up “Why?”, sent after each job or purchase, will tell you far more than a long survey few people finish. Read the answers yourself, at least at first. Patterns will appear quickly.

A few questions can help:

Which brings us back to the software company and its applauded score of 62. The more useful moment would have come after the applause: reading through the comments together, noticing that several customers mentioned faster support replies and that a few still complained about confusing invoices. The score would still have been worth celebrating. The comments would have shown the team exactly what to work on next.

Read every comment. The open question is where the real insight lives.

One Like Research & Further Reading

  1. Timothy L. Keiningham, Bruce Cooil, Tor Wallin Andreassen and Lerzan Aksoy, “A Longitudinal Examination of Net Promoter and Firm Revenue Growth”, Journal of Marketing 71(3), 2007 (abstract).
  2. Frederick F. Reichheld, “The One Number You Need to Grow”, Harvard Business Review, December 2003.
  3. MeasuringU summary of the debate, for the later balanced view that NPS and satisfaction perform comparably.

Popular Articles

Most Recent Posts

  • All Post
  • AI & Automation
  • Brand Development
  • Brand Foundation
  • Business Growth
  • Case Studies
  • Creative
  • Design Thinking
  • Digital
  • Discovery
  • Human Design
  • Insights
  • Lead Generation
  • Marketing
  • Strategy
    •   Back
    • AI Search & Discovery
    • AI Tools & Platforms
    • AI in Business
    • Prompting & Human Thinking
    • Automation & Workflows
    • Future of AI
    •   Back
    • Brand Psychology
    • Brand Experience
    • Brand Systems
    • Brand Value
    •   Back
    • Marketing Strategy
    • SEO, AEO & GEO
    • Content
    • Promotion
    • Channels
    • Performance
    •   Back
    • Human Thinking
    • Clarity
    • Innovation
    • UX & UI
    • Iteration
    •   Back
    • Trends
    • Consumer Behaviour
    • Ideas
    • Signals
    • Future Thinking
    •   Back
    • Direction
    • Positioning
    • Customer Insight
    • Growth
    • Decision-Making
    •   Back
    • Visibility
    • Startup Growth
    • Project Management
    • Systems
    • Operations
    • Execution
    •   Back
    • Marketing Discovery
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus luctus.

Products

Automated Chatbot

Data Security

Virtual Reality

Communication

Support

Services

FAQ's

Privacy Policy

Terms & Condition

Team

Contact Us

Company

About Us

Services

Features

Our Pricing

Latest News

© 2023 Created with Royal Elementor Addons

ChatGPT Prompts

About Us

Projects

Team Member

Experience

© 2026 One Like

Hello, I'm Naut!
Try our Prompt Generator to create ChatGPT workflows in seconds.