Somewhere, right now, someone is desperately trying to redeem their hard-earned points.
They’ve followed three links to get to the right page, reset the password they’d forgotten as soon as they created it two years ago, and are staring at a redemption chart, trying to decode what their points — a mere fraction of what they thought they had in the bank — can get them.
All to claim a reward that was worth maybe $15.
Was it really worth the 30-minute hassle? Probably not. Every ounce of frustration pent up in that wasted half hour is now permanently welded to your brand. Spend a few minutes on Reddit searching “loyalty points,” and you’ll see things like:
“I think loyalty programs are all a scam and are anticompetitive. Just set a fair price and let your product and service create loyalty. Stop bribing us and overburdening us with complexity.”
Here’s the funny thing, though: research shows people appreciate rewards. According to the American Bankers Association, 91% of consumers say they value their credit card rewards programs. 63% say they would be disappointed to lose that rewards program.
Below, we uncover what’s really causing this dichotomy, what opportunity that presents for marketers, and how to design loyalty programs that drive the kind of behavior you want, not the kind that gets you called a scam on the internet.
The numbers are deceptive
Per Deloitte, the average consumer enrolls in eight loyalty programs, but only participates in five. Their report states:
“With consumers actively engaging with far fewer programs than those in which they are enrolled, it seems that many loyalty programs fail to deliver sustained relevance and value.”
Loyalty programs fail to “deliver sustained relevance and value” because they were designed around what’s easy to build and measure: signups and points balances.
But if no one is redeeming anything, then you’re not building loyalty. Grant Thornton analyzed two years of transaction data for a B2C restaurant brand and found that 72% of loyalty members had low-frequency patterns. Nearly 75% of rewards went unredeemed.
Unfortunately, that’s what a “normal,” well-resourced loyalty program looks like today.
Dave Toby, Co-Founder and Managing Director of Pathfinder Marketing, a marketing agency, worked with a client that had already built its tiered loyalty structure and chosen its platform — before capturing any pre-program purchase frequency data.
Six months in, they had no way to determine whether any of it had worked.
“When you neglect that baseline, you are running a promotion that has no means of determining whether customers returned due to the program, or you are running a promotion that caused them to return when they would have been coming anyway.”
3 structural flaws of the loyalty points model
The most honest thing you can say about most loyalty programs is that they were designed for the brand’s convenience, not the customer’s.
Points are easy to issue and defer, but they come with three specific problems:
Flaw 1: Delayed value trains customers to wait for deals
One sandwich chain gave members a $25 reward after $250 in spend, and based on average customer frequency, earning that reward would have taken over two years.
Matthew Tran, founder of D2C footwear brand Birchbury, describes what happens when the purchase-to-reward gap stretches too long:
“When customers start asking you when the next promotion will be and not when you will restock, that is a sign that you have created a culture where the customer is waiting for deals and not needing the product.”
And that habit directly erodes your brand’s margins.
Flaw 2: Complexity destroys the value prop
The cognitive load of navigating a points program (downloading an app, creating an account, linking a card, tracking balances across multiple programs, activating offers before purchase) is a genuine deterrent for most consumers.
As one Redditor put it:
“Loyalty programs don’t drive loyalty. They drive mercenary behavior and are the 21st-century version of coupons and rebates in exchange for PII.”
The consumer who finds the program more trouble than it’s worth (or an invasion of their privacy) stops using it, and may even stop purchasing from the brand altogether.
Flaw 3: The economics punish both sides
Issuers carry unredeemed points as a balance sheet liability, and brands fund discounts without knowing whether the rewarded consumer was already planning to buy.
Loyalty programs may increase purchase frequency among a small set of already loyal buyers, but rarely increase the total number of buyers or materially shift category purchase frequency. Sankalp Kathuria, founder of Broadway, an experiential retail chain, explains:
“A lot of ‘retention spend’ goes to customers who were going to buy anyway — pure margin cannibalization.”
Thoughtful personalization requires first-party data
Personalization is what changes people’s behavior. To do it well, you need to know what people buy, where else they shop, how often, and at what price points.
While you can guess based on demographics, a 32-year-old woman in Brooklyn and a 32-year-old woman in Brooklyn are not the same customer. Treating them like they are is how you end up sending a vegan a steakhouse coupon.
Per Marc Bishop at Wytlabs, a marketing agency:
“Personalization means rewarding the next best behavior, not the last transaction alone. That requires first-party data on cadence, category preference, basket mix, and churn risk.”
One of the retailers his agency worked with improved repeat orders after replacing generic points with milestone perks tied to usage patterns. “My best advice? Reward one meaningful habit change.”
The hard part is getting the first-party data you need to pull something like that off. Card-linked offer platforms like Kard collect transaction-level data that can show:
- When people buy (i.e., time of day)
- Where people buy (online, in-store — with store specificity)
- What customers buy (at your store and at competitors)
- What other subscriptions your customers may have
- Where else they may shop (by category, price point, or brand)
- How much they spend, on average
Instead of blasting one-size-fits-all promos, you can segment your audience more effectively and design more appealing cash back offers, calibrated to how and when people shop. You can dream up new partnerships with brands that consumers also like. And you can play with dynamic cash back offers to boost engagement:
- Spend-tiered offers incentivize users to keep spending at your store — the bigger the reward, the bigger the purchase.
- Progressive offers give customers higher rewards for each additional purchase (2% cash back on the first purchase, 4% cash back on the second, 6% cash back on the third).
- Flash or While Supplies Last offers that can only be redeemed during a certain period of time or a certain number of times.
It also requires the right funding model
And that model is merchant-funded rewards.
Merchants cover the reward costs in exchange for exposure in fintech, tradfi, and other banking apps, helping them reach consumers at the exact moment they’re deciding where to spend. FIs and fintechs take a cut for providing the distribution channel and benefit from increased cardholder engagement. The more cardholders spend, the more everyone wins.
The best implementation of this merchant-funded model doesn’t “double charge” merchants. Meaning, they only pay if a reward is redeemed — they don’t have to pay just to be listed in the network.
That’s the way Kard operates. There are no listing fees, impression costs, or ambiguity about whether the campaign drove real behavior (read about our incrementality test process). Just to give you a taste of what this can do:
- A major food delivery platform running merchant-funded cash back through Kard brought in 10,000 new customers and reactivated 6,000 lapsed ones in a single campaign, with new customers spending at a 44% higher rate than baseline and returning customers showing 11.4% higher average order value.
- Atlas, a popular rewards credit card partner, worked closely with Kard to launch and promote cash back offers, giving cardholders instant access to cash back at thousands of retailers. Today, Atlas users who earn rewards have a 163% higher average customer value than unrewarded users, make 100% more transactions with their Atlas card than unrewarded users, and have a 22% higher average order value than Atlas cardholders who have not earned a reward.
The real test for your loyalty program
Did this specific person buy something they would not have bought otherwise? Are you sure? Can you prove it?
A points program almost never passes that test. Worse than burning margin on people who were going to buy your products anyway, a clunky points process actively pushes away people who tried to use it and then gave up.
A transaction-verified, merchant-funded program, on the other hand, is built to answer it from day one. Brands and distribution partners that are serious about making rewards a real revenue driver are replacing the ritual of handing out points with the much harder discipline of proving impact.
The next decade of loyalty belongs to the brands and issuers willing to be measured on outcomes.
