About NPS and Loyalty Measures
Introduction
I have been stumbling upon usage of NPS in different contexts, usually for gauging loyalty among customers.
I always held some skepticism, as a "one size, fits all" approach (such as the NPS score question format), usually doesn't provide any reliable measure of causation. What about correlation? Perhaps. But also in the correlation realm , there are degrees of validity. Some correlations are insightful, some are just common sense masked with fancy jargon.
Plus, let's not forget: Causation is just hard. And in marketing, scarce if non-existent.
I wanted to get an informed opinion on the subject (based on scientific evidence at our disposal on the topic).
With the help of my friendly AI assistant , I looked at evidence to come up with an answer. (So, yes , there's AI involved in the words ahead.)
Are you AI-phobic? Maybe this article is not for you, then.
Are you AI-neutral? Go ahead, read over.
In this article, I answer this simple question: should you rely on NPS score to measure loyalty?
Contents:
1. About Causation
2. About Loyalty (in Marketing) and NPS
3. Assessing validity of NPS, based on supporting evidence.
4. Loyalty and Brand Growth: The Scientific Perspective
5. Bringing It All Together
6. Back to the initial question: should you use NPS?
1. About Causation
In Marketing, Causation is hard. Why?
When it comes to marketing, inferring causation—that is, proving that a specific marketing action directly results in a particular outcome—can be extremely challenging. Marketing takes place in a complex environment with multiple variables at play simultaneously, many of which can't be controlled.
One method often used to establish causation is experimental design, wherein a control group is compared with one or more experimental groups. Changes in the experimental groups that do not occur in the control group can, theoretically, be attributed to the manipulated variables. However, conducting such experiments in a real-world market setting is difficult (if not impossible) since we simply can't control all variables.
It gets worse: it gets exponentially more hard because of:
Plus, the effects of individual marketing activities are often interrelated, making it difficult to separate the impact of one specific action.
The Ehrenberg-Bass Institute for Marketing Science (largely led by Prof. Byron's Sharp research, condensed nicely in How Brands Grow by Sharp and Romaniuk ) has contributed significant research into understanding patterns in buyer behavior and the effects of marketing actions. A fundamental principle derived from their research is the law of double jeopardy (see Sharp, Allsop for a nice review) . This law states that brands with lower market share have fewer buyers, and these buyers are less loyal. The implication is that marketing activities that aim to increase market share should ideally focus on both increasing the customer base and improving customer loyalty.
However, this doesn't necessarily imply a causal relationship between a specific marketing action and an increase in market share or customer loyalty. Rather, it highlights a consistent pattern observed across different markets and time periods. The Institute’s research has shown that changes in market share are often due to changes in the size of a brand's customer base rather than shifts in customer loyalty, but this observation doesn't identify which marketing actions are most effective at growing the customer base.
In sum, while the inference of causation in marketing is theoretically possible, practically it's fraught with challenges due to the complex interplay of variables. It's important to interpret findings from marketing research with a clear understanding of the difference between correlation and causation. Just because two variables move together doesn’t mean one is causing the other to move. The best approach often involves a combination of experimental design, longitudinal analysis, and sophisticated statistical modelling, with a healthy dose of professional judgment.
Hope that helps in thinking about causation.
2. About NPS
In today's competitive business landscape, customer loyalty is often regarded as the holy grail. Is it? Let's start by demystifying loyalty.
A definition: Loyalty is a customer's consistent preference and trust in a particular brand, resulting in repeat purchases and recommendations. It’s not just about repeated transactions, but a deeper emotional connection that turns customers into powerful advocates.
A widely accepted tool for gauging loyalty is the Net Promoter Score (NPS). It seems deceptively simple, asking customers one question: “On a scale of 0 to 10, how likely are you to recommend our company/product/service to a friend or colleague?”
But is this seemingly straightforward score really capable of predicting marketing success? Let’s dissect this.
The NPS categorizes responses into detractors (0-6), passives (7-8), and promoters (9-10). The final NPS is calculated by subtracting the percentage of detractors from the percentage of promoters. Although simple in its approach, its efficacy has become a subject of heated debate.
To make it interesting (or funny, at least to me) let's use the same simpistic scoring 1 to 10 used by NPS questions to asses validity of NPS itself.
My take (personal opinion!) based on available evidence, positions the NPS at a 5 on the validity scale from 1 to 10. Here's why.
3. Assessing validity of NPS, based on supporting evidence.
The Bright Side of NPS (5 points out of 10):
On Another note: outside of research, in the "real" world, NPS is widely (and wildly) used. So let's keep in mind that:
A. Aside from evidence of its utility, NPS got adopted by a wild array of businesses and beyond. Has proven useful in reducing complexity in strategic decision making. It's also broadly applicable. These features make the case for it being quite a nice metric.
B. Let's also remember that NPS originates from work of Bain & Company, among the titans of strategy consulting, (the company has quite good credentials , over 5 b € of revenue, and also lots of influence and incentives to popularize a method made in-house for later selling it to clients, not to mention back up their expertise in the topic). I'll leave it you the actual weight of this one factor in assessing NPS validity.
领英推荐
The NPS limits (5 points out of 10):
It's not all peaches and cream.
4. Loyalty and Brand Growth: The Scientific Perspective
NPS should help in measuring loyalty: we haven't asked why.
Why should you measure loyalty? Should you measure it at all?
Let's focus for a moment on seminal work on the topic.
Why would we want loyalty? I'd say, to make a brand grow. In other words, make it more profitable. Financials are the end goal of the marketing work, we can agree, at least in the for profit domain.
Hence, any other end usage of loyalty in Marketing, would make it a vanity metric, without implications.
But what makes a brand actually grow, according to evidence?
Does loyalty play a role?
The relationship between loyalty and brand growth is complex and nuanced. Byron Sharp's pioneering research at the Ehrenberg-Bass Institute has reshaped our understanding. Sharp contends that it's the brand's reach, i.e., the total number of customers, rather than the intensity of customer loyalty, that primarily drives brand growth.
Here's the catch: Sharp’s empirical studies indicate that customer loyalty levels do not significantly differ across competing brands within the same market. This shakes the long-held belief that deepening loyalty is a magic key to growth. Instead, Sharp proposes that brands should focus on expanding their customer base, arguing that successful brands are those with more customers who buy at roughly the same average frequency.
Evidence Based prescription for brand growth: Acquisition > Loyalty.
Also other researchers corroborated Sharp's findings.
A study conducted by Fader and Schmittlein (1993) supported Sharp's notion. They found that while customer loyalty played a role in brand choice, the size of the customer base was the critical determinant of a brand’s market share.
Furthermore, a study by Kumar et al. (2010), published in the Journal of Marketing, found that customers who were more diverse in their shopping habits and less loyal were more profitable in the long run. This challenges the focus on deepening customer loyalty and argues for a more balanced strategy of acquiring new customers and maintaining a broad customer base.
5. Bringing It All Together
While the NPS serves as a useful metric to gauge customer sentiment, it shouldn't be the sole yardstick for predicting marketing success. Its beauty lies in its simplicity, but this also becomes its Achilles heel as it fails to account for various factors influencing a customer's likelihood to recommend a brand.
Additionally, the role of loyalty in driving brand growth is more complex than traditionally believed. Recent research underscores the importance of customer base size alongside customer loyalty. The key to sustainable brand growth may lie in finding a balance between deepening loyalty among existing customers and continually reaching out to new customers.
As the quest for understanding what drives marketing success continues, one thing is clear: both NPS and loyalty are pieces of a larger puzzle. The magic happens when they're part of a comprehensive, nuanced strategy that adapts to the changing dynamics of the market.
6. Back to the initial question: should you use NPS?
Answer: it depends. What do you want to measure? for what?
Remember: customer loyalty levels don't vary significantly across brands in the same market. He argues that successful brands attract more customers who purchase at roughly the same frequency.
When deciding metrics to use, you also need to account for resources: your market research budget, company size (complexity), industry (services? consumer goods? SaaS? there are nuances), and so on and so forth.
Ultimately, NPS is:
But...
So...
If your company is using NPS, the question to be asked is: Why?
Budget constraint? Knowledge constraint? Or status quo bias?
To be sure: deep dive into the evidence. Make an informed judgement.
Ultimately, if you have the resources, concentrate efforts and budget into building better measures of brand equity, and avoid NPS.
If you don't have the resources, and currently have 0 measure of loyalty, NPS is still a good start.
NPS can help measure loyalty, even if it's not ideal based on evidence. Loyalty is just one piece of a big complex puzzle determining marketing success.
Any metric, has power when integrated in a comprehensive strategy, asking the right questions. And this part, needs to be addressed by each brand on its own.
As always ... context matters.
See also: