Loyalty programs post record membership while active engagement falls. The headline number is a vanity metric concealing the one that governs revenue.
There is a number most loyalty programs report with confidence and a number most of them cannot produce on demand. The first is membership: how many people have signed up. It goes up and to the right, it appears in board decks, and it is treated as evidence the program is working. The second is engagement: how many of those members are actually active, redeeming, and more loyal as a result. It is harder to measure, less flattering, and far more important.
In Australia, the gap between these two numbers has become a chasm. Around 86 per cent of consumers belong to at least one loyalty program, one of the highest enrolment rates anywhere, yet only about half actively engage with the programs they have joined. Enrolment is at record highs while engagement falls. The headline number says loyalty is booming. The number underneath it says something closer to the opposite.
This is not a story about loyalty programs specifically. It is a clean example of a problem that runs through brand measurement generally: the metric that is easiest to report and the metric that governs the commercial outcome are not the same metric, and reporting the first while neglecting the second produces confident decisions built on an illusion.
A vanity metric is a number that reliably goes up, looks like success, and is weakly connected to the outcome that actually matters. Loyalty membership has become the textbook case.
The mechanism is straightforward. Membership is easy to grow, sign-up friction can be lowered, incentives can be offered at the point of sale, and it produces a number that only ever increases. It is also easy to measure; the count is sitting in the database. And it looks unambiguously like progress: more members must mean more loyalty. Every incentive in the reporting system points toward celebrating enrolment.
The problem is that enrolment measures the wrong moment. Signing up is a low-commitment act, often driven by a one-time discount, and it says almost nothing about whether the customer will become more loyal as a result. A member who joined for a sign-up offer and never engaged again is, commercially, indistinguishable from a non-member, except that they cost something to acquire and they inflate a number that the organisation is using to judge the program's success.
As enrolment has been optimised, the gap between members and engaged members has widened. The result is programs reporting record membership while the share of members who are actually active erodes underneath. The number is growing and the thing the number is supposed to represent is shrinking. That is the precise signature of a vanity metric, and it is most dangerous when it looks most like success.
If membership is the wrong number, what is the right one? It is engagement that converts to commercial behaviour: members who are active, who redeem, and who are measurably more retained and higher-value as a result of the program than they would have been without it.
This is a harder number to produce, which is exactly why it is neglected. It requires distinguishing members who would have bought anyway from members whose behaviour the program actually changed. It requires linking program engagement to retention and value, not just counting activity for its own sake. And it often delivers an uncomfortable answer: that a meaningful share of the program's apparent success is concentrated in customers who were already loyal and would have stayed regardless.
The discipline here is the same one that separates real brand measurement from reassuring brand measurement. A metric earns its place in the board pack by predicting a commercial outcome, not by being easy to collect. Active, value-linked engagement predicts retention and margin. Membership predicts very little except itself. A program managed on the first is being managed; a program managed on the second is being flattered.
There is a behavioural signal worth watching inside this. Redemption patterns are a form of Elasticity Signal: the willingness to engage, redeem and respond reveals how much the program is actually shifting behaviour, and how price-sensitive and switchable the engaged base really is. A program whose redemption is shallow and concentrated is holding a small core of already-loyal customers; one whose engagement is broad and deepening is genuinely changing behaviour across the base. The membership number cannot tell these apart. The engagement signal can.
The most insidious version of this problem is that a program can be growing in membership while its genuinely engaged core is cooling, and the growth masks the cooling completely.
This is Segment Drift operating under the cover of a rising headline: the brand is losing relevance with its actually-engaged customers even as overall membership climbs. New sign-ups, many of them low-intent, replace the visible loss of engagement from the core. The blended membership number stays healthy. Meanwhile the segment that drove the program's commercial value, the deeply engaged members who redeem, advocate and resist switching, is thinning, and nobody is looking at the segment-level data that would show it.
By the time this drift surfaces in commercial results, the engaged core has often eroded substantially. The program spent its measurement attention on the growing top-line number and missed the deterioration in the segment that actually mattered. The vanity metric did not just fail to warn of the problem; it actively concealed it, because the thing going wrong was buried inside an aggregate that was going right.
Detecting this requires measuring the program at the level of engaged segments, not the level of total membership. Which members are deepening their engagement and which are drifting away. Whether the genuinely loyal core is growing or being quietly replaced by churn-prone sign-ups. Whether the program is defending the customers who matter or merely accumulating the ones who do not. These are answerable questions, but only for an organisation that has decided to measure the governing number instead of the flattering one.
Are loyalty programs themselves the problem, or just how they are measured? Primarily how they are measured. A well-designed program that genuinely changes customer behaviour is one of the most valuable assets a brand can hold. The problem is that measuring success by membership lets a program look successful without being effective, which removes the pressure to make it genuinely effective. Fix the measurement and the program design tends to follow, because the organisation can finally see what is working.
What is the single number a loyalty program should report instead of membership? There is no single magic number, but the most useful direction is incremental retention: the degree to which engaged members are more retained and higher-value than comparable non-members or inactive members. This isolates what the program actually changed, rather than counting activity that would have happened anyway. It is harder to produce than a membership count, which is precisely why it is more informative.
How does this connect to brand measurement more broadly? Directly. The loyalty case is one instance of a general failure: reporting the metric that is easy to count rather than the metric that predicts the commercial outcome. The same pattern appears in awareness metrics that do not predict consideration, in engagement metrics that do not predict revenue, and in tracking that reports sentiment rather than behaviour. The discipline is the same everywhere: measure the thing that governs the outcome, even when an easier number is available.
The loyalty illusion is comfortable because the vanity metric is genuinely reassuring. Membership goes up, the program looks healthy, the budget is justified, and no one is asking the harder question. The discomfort begins the moment someone measures engagement properly and finds that the program's apparent success is thinner than the headline suggested.
That discomfort is the point. A marketer who can show the board the governing number, how many members are genuinely active, how much retention and margin the program actually drives, and whether the engaged core is growing or being replaced by churn, is managing the program rather than being flattered by it. They can defend the spend that is working, cut the spend that is buying empty sign-ups, and catch the segment drift before it surfaces in revenue.
Given the long-standing evidence that small improvements in retention drive large improvements in profit, this is not a measurement nicety. The loyalty program is potentially one of the most powerful levers a brand has, and managing it on a vanity metric leaves that lever unpulled. The brands that measure the number that governs the outcome are the ones that turn a flattering membership chart into a program that actually defends revenue.
The membership line will keep going up. The only question is whether it is hiding a healthy program or a hollow one, and that is a question membership can never answer.
If your loyalty program reports record membership but you cannot say how much retention or margin it actually drives, you are managing a vanity metric. Brand Health designs research programs that measure the engagement and incremental retention that govern a program's real commercial value, not just the sign-up count.
Tom Morris is the Managing Director of Brand Health, an Australian brand research and brand strategy consultancy. He works with senior marketing leaders to design measurement programs that connect brand performance to commercial outcomes.