Retention & Loyalty
The other side of the coin
Friends,
I hope that all is well with you and yours as we move ever closer to what is, at least in my neck of the woods, spring. In a way, it is the most hopeful of seasons – and with all that is going on in the world, I think we deserve some hope for a change.
Before we begin, I am beyond happy to announce that I am, at long last, once more available for keynotes, lectures and presentations. Between the pandemic, travel restrictions and, most importantly, my daughter, I had limited my public speaking to only a select few occasions. Together with some absolute corker decks, not least the one below on adaptive strategy which features my favorite opening slide reference to date, I am now ready to make audiences sit upright again:
For any inquires into my availability, whether for physical or digital talks, just send me an email.
To the business at hand
A week ago, we discussed the basics of customer acquisition; buyer distributions, costs, promotions and, as it turned out, weddings. Today, we are going to continue the conversation with an introductory look at retention. Next week, for paying subscribers, we will conclude the series with a deep dive into nuance – subscription markets, platform companies and metrics.
But that is then and this is now, so first things first.
Although acquisition and retention often are handled separately, they are intrinsically linked; overall expansion of the customer base can, logically, come both from increasing the influx of new buyers and decreasing the rate of departure of old ones. This leads to a number of critical questions, such as:
What carries the best cost-benefit ratio – acquisition or retention?
What are the implications for loyalty?
What are the implications for measures such as CLV?
So, let us attempt to answer them.
The cost-benefit conundrum
Given that company resources are finite, strategic decision-making habitually comes down to questions of financial prioritization based on either a) the potential upside of a path, or b) the potential downside of alternative paths. It is hardly surprising, therefore, that the retention narrative historically has featured both. As many no doubt will be aware, among the most widely repeated claims are that a 5% increase in retention would boost the company profits by as much as 100%, and that it would cost five times as much (or more) to acquire a customer than to retain one.
Yet upon customary critical inspection, it turns out that neither assertion is rooted in reality.
Bain consultant, NPS creator and loyalty enthusiast Frederick Reichheld was, together with Harvard professor Earl Sasser Jr., among the first to argue that a seemingly minor improvement in churn could massively increase a company’s fortunes. In the grandiosely titled Zero Defections: Quality Comes to Services, the pair held, for example, that retention rates were ‘an accurate leading indicator of profit swings’ and that ‘longer term customers generate increasing profits’.
However, even though the authors were only too happy to make comments such as ‘understanding the economics of defections is useful to managers in several ways’, it became apparent that they themselves struggled with the basics. As any reader who stuck with the article beyond the first few paragraphs would come to realize, it was not company profitability that increased but customer profitability, an altogether different thing.
To make matters worse, the eyebrow-raising 5% contention referred not to five percent, but five percentage points. Lowering churn from 10% to 5% is, as most with a rudimentary understanding of mathematics should know, not a 5% difference but a 50% difference. Even with fake data created to prove a point, they were off by a factor of ten.
But the oddities did not stop there. In the imaginary scenario provided, it was assumed that improvements made would be achieved at zero cost. This, of course, is yet another nonsense; in reality, the investments needed to halve defection rates would be substantial.
The precise origin of the ‘five times’ claim, meanwhile, is more difficult to pinpoint. By my best estimate, it appears to have come out of research conducted by the Technical Assistance Research Project (TARP) in the late 1980s.
At the core of this argument was - stop me if you have heard this one before - a number of assumptions. In particular, it was taken for granted that customers would
increase their spend over time,
purchase more at full price rather than at discount prices,
create operating efficiencies for firms (as the two parties became familiar with one another), and
increasingly recommend the company to others.
These assumptions are often seen in loyalty discourse; Reichheld and Sasser repeated them, as have many others. Unfortunately, while each hypothetically could be true in any given context, subsequent research has demonstrated that it is highly unlikely.
We know from the law of buyer moderation that there is no guarantee that customers become more profitable over time. In fact, the relationship between value and tenure inherently cannot be linear as every customer relationship inevitably will end (people, like companies, die). The group most likely to buy at discount prices are those who already purchase the brand. Existing buyers are rarely cheaper to serve (in fact, they have repeatedly been shown to be more expensive to serve), and referrals tend to conversely decrease over time due to initial honeymoon effects.
In other words, as much as it might seem plausible that acquisition costs would be significantly higher than retention costs, the truth is as ever that the corporate reality is significantly more complex and context dependent than the pervasive story would have you believe. Following what one might call conventional wisdom or universal induction can thus lead to anything from suboptimally allocated resources to financial calamity.
And yet, the demise of loyalty is greatly exaggerated
When assumptions underlying a theory are shown to be erroneous, it stands to reason that any subsequent reasoning built upon them becomes fatally flawed. Provided what we have just unearthed, does this mean that retention activities are meaningless or that there would be no point to things such as loyalty?
No, of course not. Even though similar conclusions are occasionally drawn – not least by those who have read How Brands Grow somewhat sloppily – loyalty is still a key factor in business. But it is clear that we need to reconsider the traditional view of it.
This requires, firstly, that we come to terms with what the phenomenon actually entails. Although there are many ways in which to define loyalty (e.g., share of category requirements), the everyday language interpretation of it being a form of faithfulness to a single entity is rarely representative of buying behavior. Rather, buyers tend to be loyal to a repertoire of brands (in line with their penetration, as per duplication of purchase law). Adding the previously highlighted double jeopardy law, which states that smaller brands have fewer customers (the first jeopardy) and that those customers also are slightly less loyal (the second jeopardy), means that while improving loyalty is not out of the question, it is most easily achieved through increased market share. Or to put it slightly differently, through acquisition, not retention.
That is not to say that retention would be a pointless endeavor. Certainly, it can be justified, but defection rates will be difficult to improve beyond what is natural given the relevant context, and the financial merits thereof should therefore be carefully considered.
But what about Customer Lifetime Value?
The counterargument to the above is that retaining customers nonetheless can be a prerequisite for business longevity; particularly in tech-driven subscription markets, it is not unusual that companies initially make a loss on every new customer. Getting them to repeat their purchases and increase the value to the company is thereby the only viable path to profit.
But while that undoubtedly may be the case, relying on customer lifetime value (CLV) to provide the balance (as most acquisition vs retention models do) is a much tricker proposition than many realize.
As a rule, CLV is calculated in a static ‘now’, that is, the net present value (NPV) of all anticipated cash flows over a given time span for a given customer. A positive NPV means an attractive customer, a negative NPV is a reason to stay away. But although I do not object to the practice of allocating budgets to segments according to probable future cash flows per se (to the degree that one can know such things), this can easily become an exercise in targeting the most valuable customers. As we know, there are neither guarantees that heavy buyers will increase or even maintain their spend, nor that retention would be less expensive than acquisition. Buyers who have a negative NPV at the time of the calculation may also become highly profitable over time – and unsurprisingly, this is often the case.
Consequently, making blanket statements about CLV based on isolated observations in the present can be, and often is, limiting to future earnings. Developing a successful acquisition and retention approach instead requires a longer-term, holistic perspective - and the knowledge that the link between loyalty and profits often is weaker than commonly expected.
Indeed, at the end of the proverbial day, none of the usual justifications for investing in loyalty stands up particularly well to practical examination. That does not mean that there would not exist others, but as I have hopefully demonstrated today, discovering them requires an in-depth understanding of the business model used and insight into the dynamic nature of buying behaviors.
Next week, in a paying-subscriber exclusive, we are going to look at the exceptions and applications, particularly for subscription markets.
Until then, have the loveliest of weekends.
Onwards and upwards,
JP