Brands spend enormous amounts of money creating incentives designed to influence customer behavior. Points, rebates, promotional credits, cashback, gift cards, loyalty currencies, coupons and offers all operate somewhat differently, but economically they’re trying to accomplish much the same thing. A business creates something of value and gives it to a customer with the expectation that the value will contribute to a desired outcome — a purchase, another purchase, a larger basket, a renewal, a return visit, a referral or some other measurable behavior.
We’ve gotten pretty sophisticated at the front end of that process. There are entire technology categories devoted to determining which customers should receive an incentive, how much it should be worth, when it should be delivered and which behavior it’s intended to produce. Considerable effort also goes into measuring what happens afterward. We know how much value was issued, how much was redeemed, what it cost and, with varying degrees of confidence, what revenue resulted.
What’s less well understood is what happens in between.
There is a fairly consequential distance between issuing an incentive and a customer realizing its value. Inside that distance are all kinds of opportunities for the value to get lost, ignored, misunderstood or simply forgotten. We tend to roll much of this up under the heading of redemption, but doing so obscures an important part of how incentives actually perform.
And it can make redemption rate a fairly misleading measure of success.
Redemption friction starts well before redemption
The obvious version of redemption friction is procedural. Too many steps. A promo code that doesn’t work. A forgotten password. An offer that can’t be used in the channel the customer happens to be shopping in. Those are easy to recognize because the customer has already decided to redeem the value and something is standing in the way.
But a great deal of redemption friction occurs before the customer gets anywhere near that point.
The customer may not know the value exists. They may have seen it when it was issued but forgotten about it. They may know they have points somewhere without knowing what those points are worth. A reward may be sitting in an app they rarely open, a loyalty account they don’t regularly visit or an email that disappeared beneath several hundred other emails. There may be a minimum purchase requirement, an expiration date, an eligible product set or some other condition that isn’t apparent until the customer tries to use it.
All of those things affect redemption, but they aren’t really redemption problems. They’re visibility, comprehension and access problems.
That difference becomes important when we start trying to understand why incentives perform the way they do.
Take a simple example. Two customers each receive a $25 incentive. One sees it, understands it, knows where and how to use it and decides it isn’t compelling enough to act on. The other never notices the email announcing that the incentive was issued.
Both incentives go unredeemed. They will probably be recorded and reported the same way. But only one of them has actually told us anything about whether $25 was sufficient to influence the customer’s behavior.
The other never really entered the equation.
What does redemption rate actually measure?
Redemption rate is useful. It’s just being asked to tell us more than it can.
If 100,000 offers are distributed and 28,000 are redeemed, we know something important about those 28,000. What we know about the remaining 72,000 is considerably less precise.
Some customers weren’t interested. Some never saw the offer. Some saw it and forgot about it. Some couldn’t find it when they wanted to use it. Some encountered restrictions. Some allowed it to expire without intending to. Others may have looked at it carefully and decided the value simply wasn’t worth the effort.
Those aren’t variations of the same outcome. They are different outcomes with different implications for program design, customer experience and economics.
Yet the incentive industry has traditionally had much better visibility into issuance and redemption than into what happens between them. We know when value enters the system and when value leaves the system. The middle is considerably murkier.
A more useful view of the incentive lifecycle might look something like this:
Value funded → Value issued → Value discovered → Value understood → Value accessed → Value redeemed → Behavioral outcome
Look at the process that way and redemption starts to look less like the central event and more like one stage in a longer conversion funnel. There can be leakage at every point, and the reasons for that leakage aren’t interchangeable.
Breakage makes the picture even murkier
There is also an economic wrinkle here that makes incentives somewhat unusual.
Unredeemed value isn’t always treated as a failure. Depending on the structure of the program, breakage can reduce the actual cost of an incentive or ultimately produce an accounting benefit. That has created an understandable tendency to view some level of non-redemption favorably.
But it raises an awkward question. Why was the incentive created in the first place?
Presumably, the business wasn’t issuing $25 because it wanted to save $25. It was issuing $25 because it believed that value could influence customer behavior and produce an economic return greater than its cost.
Seen through that lens, an incentive that goes unused because the customer consciously rejected it is very different from one that goes unused because the customer forgot it existed. The first may tell us something about the economics of the offer. The second may tell us something about the effectiveness of the system delivering it.
There is a useful distinction to be made here between financial efficiency and behavioral effectiveness. Breakage may improve one while undermining the other. At the very least, we shouldn’t assume that unredeemed value is economically productive simply because it wasn’t ultimately paid out.
Measure the journey, not just the redemption
A better measurement model would start by treating redemption as a funnel rather than an event.
How much issued value was actually visible to the customer? Of the value that was visible, how much was understood? Of the value that was understood, how much was readily accessible? And once the customer had a realistic opportunity to use it, how much was redeemed?
This starts to separate customer choice from system friction.
It also gives businesses a much better diagnostic framework. If customers see an incentive, understand it, can access it easily and still don’t use it, the offer itself may be the problem. If customers who encounter the incentive redeem at very high rates but overall redemption remains low, increasing the value of the incentive probably isn’t going to fix much.
Some of the information needed to make these distinctions already exists. Message delivery and engagement data can tell us something about visibility. Account and wallet activity can tell us whether customers are encountering their value. Expiration patterns can expose value that is consistently being lost. Customer service inquiries, failed redemption attempts and abandoned transactions can reveal access problems.
None of these measures is particularly exotic. What’s missing is the idea that they are all describing different parts of the same value-realization process.
More importantly, redemption shouldn’t be the end of the measurement exercise. Incentives aren’t generally created for the purpose of generating redemptions. They’re created to generate behavior. The meaningful economic measures are incremental spend, frequency, retention, reactivation, category adoption, basket size or whatever behavior the program was designed to influence.
A much more interesting performance metric might therefore be behavior influenced per dollar of incentive value created.
The solution isn’t necessarily a better incentive
Once we look at incentives this way, the solution becomes broader than improving the redemption experience.
The objective should be to increase the percentage of incentive value that gets a legitimate opportunity to influence customer behavior.
That starts with visibility.
Most incentive systems are very good at recording that value exists. They’re considerably less effective at making sure the customer knows it exists. An incentive sitting inside an account ledger is technically available but behaviorally inert until the customer encounters it.
Visibility therefore needs to become an operating requirement rather than a communications afterthought. Customers should encounter available value in places where they’re already making decisions — during shopping, inside account experiences, in wallets, in payment environments and at other moments when that value is actually useful.
Timing matters. A $20 credit discovered three weeks after a purchase is a nice surprise. The same $20 discovered while the customer is deciding whether to make the purchase may be commercially much more powerful.
The next problem is comprehension.
The incentive industry has created an extraordinary number of currencies, constructs and rules. Customers encounter points, miles, stars, credits, bonus dollars and other units whose actual value often requires some amount of translation.
There may be perfectly good reasons for those constructs. But there is also a cost whenever the customer has to do the math.
A balance of 14,700 points may have immediate meaning to a highly engaged program member. “$147 available” requires considerably less interpretation.
Programs don’t necessarily need to abandon their currencies. Customers just shouldn’t have to become experts in those currencies to understand the value they possess.
And then there’s access.
If using an incentive requires finding an old email, remembering a password, downloading an app, locating a code and determining whether the purchase qualifies, every additional step creates another opportunity for the value to become irrelevant.
A useful test might be: how much work are we asking the customer to do to receive the benefit we told them we’d already given them?
The customer doesn’t see a program
There is another complication that becomes easy to overlook when incentive performance is measured program by program.
Customers don’t experience incentives that way.
A person might have airline miles, hotel points, retailer rewards, cashback balances, promotional credits, gift cards, card-linked offers and expiring benefits spread across any number of programs. Each one may be well designed. Each may have a perfectly good app, account experience and communications strategy.
But none of them operates in isolation from the customer’s point of view.
Every program is competing for a share of the same limited attention. Every new app, account, email, balance and expiration date adds another small piece of administrative responsibility for the customer. At some point, the cumulative burden becomes part of the friction.
This is worth considering because it means redemption friction isn’t always created by a bad redemption experience. It can also be created by the sheer volume and fragmentation of customer value in the marketplace.
A brand may have made its reward relatively easy to use once the customer arrives at the right place. The larger challenge may be getting the customer to remember that the reward exists at the moment when it could actually matter.
That puts greater importance on how incentive value is communicated and surfaced over time. A notification at issuance isn’t necessarily enough. Neither is making a balance available somewhere inside an account. The value needs to remain discoverable and understandable throughout its useful life, particularly at moments when it has the potential to influence a decision.
There are plenty of ways to improve that today. Better integration with existing customer touchpoints. More useful balance and expiration communications. Clearer expressions of monetary value. Fewer unnecessary steps between awareness and use. Greater consistency across channels. Smarter use of wallets, apps, payments and other environments where customers are already interacting.
The specific answer will vary considerably by program. The principle shouldn’t.
If an incentive was created to influence behavior, the system surrounding it should be designed to maximize the probability that the customer actually encounters that value while it can still do so.
That’s a different standard from simply making the value available.
Make the value you already create work harder
The incentive industry has traditionally responded to performance challenges by adjusting the offer: change the audience, increase the value, change the threshold, improve the creative, alter the timing.
Those are all legitimate levers.
But if the underlying problem is friction, increasing the incentive can be an expensive way to solve the wrong problem.
Imagine that a $20 incentive is highly effective among customers who actually encounter it, but a large percentage of recipients never do. Increasing the offer to $25 or $30 doesn’t address the failure. It simply puts more money behind the same delivery problem.
The better investment may be in making the original $20 more visible, understandable and usable.
That creates a different optimization question. Instead of asking how much additional value needs to be created to produce another unit of behavior, businesses can ask how much additional behavior can be generated from value they’re already creating.
There’s potentially a lot of leverage in that distinction.
The incentive industry has invested heavily in getting value to customers. Targeting is better. Personalization is better. Funding is better. Issuance is better. Attribution is better.
The next opportunity may be making sure that value survives the trip.
Because the interesting number isn’t simply how much incentive value was redeemed or how much went unused.
It’s how much of the value we created was actually given a fair opportunity to do the job we created it to do.
Photo by Trebyx Consultoria on Unsplash
