Points, Perks, and Broken Promises: The 4,000-Year Trap of Loyalty Programs

Somewhere in your wallet, your phone, or your email inbox, there are points you will never redeem. Maybe they're airline miles with blackout dates that make them functionally useless. Maybe they're coffee shop stamps on a card you've lost three times. Maybe they're tokens in a retail app that requires a software update before you can even open it.

You are not uniquely bad at this. This is not a you problem. This is a 4,000-year-old trap that has fooled customers and eventually destroyed companies with remarkable consistency — and yet every decade, someone reinvents it and calls it innovation.

The Oldest Loyalty Scheme We Know About

Around 1800 BCE, Egyptian merchants operating near major temple complexes began offering preferential treatment to repeat customers — early access to goods, discounted rates, and symbolic tokens that could be exchanged for future purchases. We know this because temple administrative records, some of the most detailed economic documents from the ancient world, tracked these arrangements with obsessive care.

The system worked. For a while. Customers returned. Relationships deepened. The merchants who offered the most generous terms attracted the most business.

Then the merchants discovered what every loyalty program operator eventually discovers: the cost of the rewards started eating the margin. Customers who had been attracted by generous perks expected those perks to continue or expand. Scaling back triggered defection. The merchants who had used rewards to build their customer base found themselves trapped — they couldn't afford to maintain the program, and they couldn't afford to end it.

Several of the temple records document merchants who went under specifically because their customer retention schemes became financially unsustainable. The archaeologists who translated these documents in the 20th century noted, with some amusement, that the failure pattern was essentially identical to what was happening to American retailers at the time.

Rome Tried It Too

Roman trade guilds — the collegia — developed sophisticated retention mechanisms for both members and customers in the first and second centuries CE. Guild membership came with access to collective buying power, shared legal protection, and social benefits like funeral assistance. For customers, guilds maintained preferred-customer registers that entitled repeat buyers to price stability and priority access during shortages.

These weren't informal arrangements. They were contractual, recorded, and enforced. Roman merchants understood, intuitively, that a customer who had invested in a relationship with a guild was less likely to defect to a competitor.

The problem — and historians of Roman commerce have documented this across multiple regions — is that the benefits promised to preferred customers had a habit of quietly degrading over time. The price stability became less stable. The priority access came with more exceptions. The guild kept the customer's loyalty investment while gradually delivering less in return.

When customers finally noticed — and they always eventually noticed — the backlash was disproportionate to the original grievance. It wasn't just that the deal had gotten worse. It was that they felt deceived. The trust damage was worse than simple price competition would have been.

That pattern — build loyalty through rewards, quietly devalue the rewards, lose the customer in a trust collapse rather than a price war — is so consistent across the historical record that it almost functions as a law.

The Modern Reinvention Cycle

American Airlines launched the first modern frequent flyer program in 1981. It was genuinely innovative — a systematic, scalable way to reward repeat customers with something that felt valuable. Other airlines copied it within months. Hotels followed. Credit cards followed. Eventually, grocery stores, coffee chains, gas stations, and pizza delivery apps all had their own versions.

By the early 2000s, researchers estimated that American consumers held more than $48 billion in unredeemed loyalty points. That number has only grown. A 2023 analysis put the global value of unredeemed loyalty currency at over $360 billion.

Here's what that number actually represents: it's the gap between what companies promised and what customers ever received. It's the accumulated weight of a broken deal, sitting on corporate balance sheets as a liability that companies hope customers will simply forget about.

Most of them do forget. Which is, of course, the entire business model.

NFTs, Subscriptions, and the Latest Reinvention

In the early 2020s, a wave of companies discovered that blockchain tokens could function as a new kind of loyalty program. NFT-based membership gave holders access to exclusive content, early product drops, and community spaces. The pitch was that unlike traditional loyalty points, these tokens could be traded — they had real-world value that the company couldn't unilaterally destroy.

By 2023, most of these programs had collapsed. The tokens were worth fractions of what early buyers paid. The exclusive access turned out to be access to things nobody particularly wanted. The communities dispersed.

The subscription perk model — pay monthly, receive benefits — has had a somewhat better run, but the pattern is visible: services launch with generous perks, quietly reduce them as subscriber growth slows, then face cancellation waves when customers finally notice the deal has changed.

Amazon Prime has raised its price four times since 2014. Each price increase was followed by analysis showing that most customers stayed anyway because they'd already built their behavior around the service. That's not loyalty. That's switching cost masquerading as loyalty. They're not the same thing, and the historical record is very clear about which one is durable.

What Actually Keeps Customers

This is where 4,000 years of data becomes useful, because the historical record is not purely a chronicle of failure. There are businesses — ancient and modern — that maintained customer relationships across generations without elaborate reward schemes.

The consistent variables aren't complicated: price fairness, product consistency, and the absence of surprise. Customers who felt they were getting a square deal, every time, without hidden conditions or gradually shifting terms, stayed. Not because they were locked in by sunk costs or dazzled by points. Because they trusted the transaction.

Roman merchants who maintained stable prices during shortages — when they could have gouged — built customer relationships that outlasted their competitors who had more sophisticated retention schemes. Medieval guild records show that the craftsmen with the best long-term customer retention were almost universally the ones with the fewest complaints about product quality, not the ones with the most generous membership perks.

Modern research confirms this consistently: the strongest predictor of customer retention isn't reward program participation. It's complaint resolution. How a company handles the moment when something goes wrong determines loyalty more reliably than any points structure.

The Program That Keeps Getting Rebuilt

None of this stops companies from launching new loyalty programs. The appeal is too strong — rewards feel like a controllable lever, a way to manufacture the retention that should come from simply being good at what you do.

And in the short term, they work. Enrollment spikes. Engagement metrics improve. The quarterly numbers look good. The problems arrive later, when the cost of the program becomes visible, when the rewards get quietly trimmed, when the customers who joined for the perks realize the perks were always worth less than advertised.

At that point, the company faces the same choice Egyptian merchants faced 4,000 years ago: maintain the program and bleed, or cut it and watch the trust collapse.

The historical answer is usually some version of both, in the worst possible order.

Maybe the lesson isn't that loyalty programs don't work. Maybe it's that what they're designed to manufacture — trust — is the one thing you can't manufacture. You can only earn it, slowly, by being fair and consistent when it would be easier not to be.

That's not a great pitch for a product launch. But it has a significantly better track record.