I am regularly asked for my opinion of a performance metric called the Net Promoter Score (NPS), which has become a popular way to assess an organization’s effectiveness by measuring the ratio of loyalists to detractors. Many companies have adopted the metric as a leading indicator of growth — the logic being that as the NPS score goes up, increased growth will naturally follow. NPS is often adopted at the urging of very senior levels of management, and people want assurances that it’s worth the effort. My advice is always to not take anyone else’s word for its utility; test NPS with your organization’s own data, and if NPS correlates with growth (or whichever ultimate measure of performance you are trying to improve), consider using it.
A pattern I’ve observed is that many firms adopt NPS because of the rationale behind it — not because the evidence at their organizations supports its use. Indeed, an uncomfortably large number of managers have sheepishly admitted to me that they found no correlation between their NPS scores and ultimate performance, long after they’ve adopted NPS as a key organizational metric. This puts them in a terrible bind. NPS is often embraced with great fanfare, complete with incentives and systems that are redesigned to focus the organization on improving the metric. When NPS is not a leading indicator of growth, organizations expend tremendous effort with little to show for it.
After seeing enough of these examples, I began to investigate what was going on. Here’s what I’ve learned. The logic of NPS is very persuasive. It’s clean and intuitive, and feeds our hunger for one clear goal on which we can focus the troops. If a company has more advocates than detractors, good things should happen. And the greater the imbalance, the better the things, right? The problem is that while this might be true as a general principle, it may not be true in a specific context. For example, if all those advocates are consuming your outstanding pre-purchase service, but closing the deal with cheaper competitors, as happened to Gateway computers, their advocacy is doing you little good.
Here’s my advice. Most organizations are already collecting the big three customer metrics — customer satisfaction, customer intention to repurchase, and customer likeliness to recommend. I strongly encourage organizations to have a runoff between all three measures and the performance measure they’re trying to improve. If NPS is the strongest driver of performance, either in an existing time period or as a leading indicator, then by all means use it. But if it’s not, I suggest discarding it, regardless of the shiny appeal of its high-level logic.