Popular legend says marketing today is living on the edge of chaos: the new heroes are from Facebook and Google, ‘link bombing’, ‘paid buzz’ and ‘friend poking’ are hotly debated, and the new lexicon includes ‘authentic’, ‘social network’ and ‘word of mouth’ (WOM). Out of this chaos, Net Promoter1 has emerged as the true compass to guide the new age marketing traveller.

The aim of this paper is to stand back from the breathless chatter and reflect on the significance of what's happening. Net Promoter is a nonfinancial metric that should be seen in the broader context of performance measurement and management. It is just one example of many metrics that have been sold to managers hungry for the latest new thing. Studies by Bain & Company2 and by Richard Pascale3 show that managers have a big appetite for the latest fashions, and adopt one or more new tools a year, only to discard them within about 18 months.

Metrics came back into fashion a decade ago, when Kaplan and Norton wrote the Balanced Scorecard,4 a reinvention of the earlier key performance indicators that had been fashionable ten years earlier. They defined some important ground rules concerning the use of metrics in scorecards and dashboards, which have been widely accepted and are important to any assessment of the suitability of Net Promoter as a useful management metric. Here's what they say are the essential elements of an effective scorecard.

‘The best Balanced Scorecards will tell the story of the strategy so well that the strategy can be inferred by the collection of objectives and measures and the linkages among them…The strategy story is told by linking outcome and performance driver measures via a series of cause and effect relationships. The outcome measures tend to be lagging indicators. They signal the ultimate objectives of the strategy and whether near-term efforts have led to desirable outcomes. The performance driver measures are leading indicators, which signal to all organisational participants what they should be doing today to create value in the future…The essential objective in selecting specific measures for a scorecard is to identify the measure that best communicates the meaning of a strategy. Every strategy is unique’.

Therefore, against these widely accepted views about choosing measurements, let's look at Net Promoter and ask the following questions:

  1. 1

    Is Net Promoter a lead measure?

  2. 2

    Does Net Promoter have the right characteristics in terms of cause and effect?

  3. 3

    Is Net Promoter a general purpose measure that most organisations can and should adopt?


According to Kaplan and Norton, customer metrics like Net Promoter ought to be lead indicators that precede changes in financial indicators like revenues. Reichheld cites a study conducted by the Listening Company in conjunction with the London School of Economics in which they demonstrate correlation between financial results in 2003–2004 and Net Promoter scores in 2005. Therefore, on this evidence it's a lag indicator — people report high Net Promoter scores a year or more after growth occurs. One interpretation of this is that they are reacting to the apparent popularity of the products; they recommend what they perceive to be growing products.


According to Kaplan and Norton, good indicators are linked to cause and effect relationships. Reichheld does not attempt to demonstrate this himself, but there is ample literature on consumer behaviour that provides useful clues.5, 6

Involvement is a fundamental aspect of consumer behaviour, and low-involvement purchasing behaviour is utterly different from high-involvement purchasing. The vast majority of purchases are low-involvement and habit is often the guiding force for the consumer. These low involvement purchases by their nature don't generate WOM buzz and therefore Net Promoter can be ruled out as a useful measure on cause and effect grounds.

High-involvement purchases are items that are important to the purchaser and are closely tied to their ego and self-image. Cognitive activity and extended problem solving occur when the purchase is more involving and WOM is more likely. WOM, however, still has a peripheral impact even for high-involvement purchases.

There is extensive research on ‘where did you hear about our product’ collected by thousands of companies, and WOM seldom scores more than a few percent even in high-involvement product areas. Take, for example, pensions and motorcar purchasing, both of them high-involvement purchases. WOM is only a minor factor in pensions because the main source of advice is the Financial Advisor. The brand you choose is unlikely to coincide with dinner party chatter, even if you have a vocal acquaintance who talks about their pension — most people value the advice of their financial advisor more. In the case of motor vehicles, although people talk a lot about their cars and likes and dislikes, the final purchase decision is a highly personal one and more likely to be influenced by reviews in magazines or on television. Ask yourself, out of all the thousands of brands you have ever purchased in your entire life — clothes, shampoo, wrist watches, beer, insurance, car, cosmetics — what percentage of your brand and product choices were specifically copying one of your friends' choices of products and brands?

A very small number of examples are cited over and over again by WOM protagonists. For example, the popularity of the pop artist Lily Allen, who promoted herself via social networking rather than more traditional media. Yet this is precisely the type of ultra-high involvement purchase where WOM would be expected to be important, and as an example it's very unrepresentative of the other purchase categories.

Thus, cause and effect analysis of consumer behaviour indicates that (1) for most products and services there is little WOM and (2) even where WOM is more commonplace, people aren't necessarily influenced very much by it, except in very rare situations.


Kaplan and Norton give general guidance to anyone selecting metrics. They say ‘your strategy is unique and so don't expect to find general purpose metrics’. Net Promoter's claims fly in the face of this advice: that they are a general purpose metric that fits all situations.

Following the logic of the previous section about high-involvement and low-involvement situations, it would appear that Net Promoter can only be expected to have any relevance at all in very high-involvement situations, where there are obsessive fashion followers.

What is strange then is why Reichheld's book has had many 5-star rankings on Amazon.com. As one reviewer7 notes, however, ‘most of the 5-star rankings can be attributed to BzzAgent. Doesn't it seem ironic that someone espousing the power of recommendation would enlist paid recommenders to promote his book?’

Business Week8 took up the story and this is what they had to say: ‘One strategy it [Reichheld's book] doesn't suggest, however, is launching a word-of-mouth campaign with a firm that has an army of ready-to-buzz folks equipped with free books, background material, and guides that include ideas for conversation starters. Yet when marketing his book, Reichheld did just that. As at least one rival in the customer satisfaction field has noted, 12 of the book's 40 reviews on Amazon.com were written by self-identified members of BzzAgent, a Boston-based word-of-mouth marketing firm. Nearly all 12 were glowing…Reached later, Reichheld said… “I thought since I was working in this area of word-of-mouth it would almost be irresponsible of me not to take that experiment”’.

As legal expert, Anna Montes at Osborne Clark has commented to The Marketer:9 ‘The Consumer Protection from Unfair Trading Regulations 2007 will cover more than the aggressive sales techniques…Falsely representing oneself as a consumer will be a criminal offence under the regulations’.

So where does this leave Net Promoter? Well, it seems to be a lag indicator, whose cause-and-effect pedigree is poor and which is unlikely to have any serious management value in any except very high-involvement purchases where people follow the advice of friends and acquaintances. In those situations, do choose Net Promoter, but apart from those I'd suggest you ignore it.