I've had to cut program benefits, and it was awful. We had to roll back where we offered the program and its perks geographically, and we didn't replace what we took away. I knew it was coming and that it had to happen for the sake of the business, but I didn't want to do it. We grandfathered members in for a grace period to wean them off, and it still stung. Giving and improving benefits is always wonderful, and dealing with the fallout when you take away something people like is the absolute worst.

A lot of loyalty teams are about to face the same call. By now, most 2027 budgets are set or close to it, and teams are walking into the new year knowing they’ll have to act on those decisions soon. Gartner’s 2026 CMO Spend Survey found marketing budgets rose only slightly, to 7.8% of company revenue, and 56% of CMOs said they don’t have the budget they need to deliver their strategy [1]. When the budget comes in flat and the expectations don’t, loyalty program benefits are one of the first places to give, which is why I want to talk about the order in which you make these changes.

Sometimes cutting rewards is the right call. Every point you issue is money you owe members later, and a program that gives away more than it earns back isn't helping anyone. The same goes for discounts and rebates, because it’s all deferred value. But a change that makes rewards less valuable to members, which the industry calls a devaluation, is something members take personally. And that’s an important consideration. The public examples that went badly have a few things in common, and I’m sharing them here so you don’t repeat them. The brand cut in the wrong order, said too little about what was coming, or took away something members felt they had already earned.

And if you’re a consumer brand, there’s one more thing you should think about. When an airline or a hotel cuts benefits, their members go online and complain, so the brand hears about it right away. Your members probably won’t do that, at least not in a way you’ll easily see unless you’re leaning into social listening. So the impact starts out subtle and shows up later, usually in the next quarter’s review. So the way I look at it, how you make the change and how you measure it matter almost as much as what you decide to cut.

Expedia shows why order matters

Expedia's One Key program is a good example to learn from because it’s recent, and even though it’s a travel program, the lesson applies just as much to a consumer brand. In June, Expedia announced that starting July 28, members in its entry tier would earn 1% back on hotels, packages, and other non-flight bookings instead of 2%, and it dropped its hotel Price Guarantee [2]. Then in October it added new perks, including lounge access when a flight is delayed at least 90 minutes, money back when a fare drops, and a discount on CLEAR [3][4].

On paper, that seems like a reasonable trade, and I can see why they made it. But the new perks came about ten weeks after the cut and most of them went to the top tiers, and the new price protection only covers flights for Gold and Platinum members [3]. One travel blog summed up the reaction for entry-level members by saying there wasn't really anything there for them [5]. So by the time the new benefits showed up, many members had already felt the cut and the frustration, and the new perks weren't built for most of them.

The sting of loss is stronger than the positive effect of earning status

The most useful research on this comes from a study of loyalty status demotion. Researchers found that the negative effect of losing status was stronger than the positive effect of gaining it. In a field study with a European department store chain, demoted customers spent 17.9% less and made 21.5% fewer transactions than in the same period the year before [6].

The same paper also ran a separate experiment, set in an airline program, to see what softens the blow. Explaining the membership conditions and showing customers their own spending reduced the negative reaction, and so did an apology. Offering money to make up for the demotion had no effect. None of these fully canceled out the damage, but they helped [6].

Delta learned this publicly in 2023. It announced much higher requirements to earn elite status, long-time members pushed back hard, and about five weeks later it rolled most of the changes back [7]. Delta's CEO said the changes "probably went too far" [8].

So if you have to cut, start with how members earn going forward, and leave the balances and status they already hold alone for as long as you can. The CFPB made a similar point about credit card rewards in late 2024, warning that reducing the value of rewards already earned without clear explanation or adequate timelines may be unlawful [9]. That guidance is aimed at card issuers, but the principle is a good one for any program.

Which loyalty program benefits to cut, and which to add

Look at these first when you need to cut cost

• Earn on low-margin categories or channels.
• Rewards almost nobody redeems.
• Blanket discounts that mostly go to members who would have bought anyway. If you’re not sure how to tell, my post on measuring program lift walks through it.

That last one is usually the biggest. A poorly designed program can reward behavior that would have happened anyway, turning full-margin sales into discounted ones without adding any value [10]. I walked through the math on that in my post on holiday discount codes, and why discounts don't make fans covers what they do to loyalty over time. And as I wrote in this post on the most profitable customer, that customer is often the one who never needed the offer in the first place.

If you're weighing loyalty program costs more broadly, that post covers when a platform pays for itself and when it doesn't.

Add perks that cost little and work on their own

• Early access to launches, restocks and sales. Bond's 2026 Loyalty Report found that access, meaning first looks, last calls and getting to the front of the line, was the top driver of loyalty in both 2025 and 2026 [11].
• Service perks, like a longer return window or a priority support line.
• Automatic protections, like a price adjustment if something goes on sale soon after purchase.
• More time to use what members earn. Antavo's 2026 research found that 41.1% of consumers said rewards expire before they can use them [12]. Chipotle made this change in its 2026 relaunch. Points now stay active as long as a member makes one qualifying purchase a year, up from every six months [13].

EY's 2026 Loyalty Market Study found programs moving past discounts and points toward membership-style benefits, with access becoming a major design theme. Member-only events or experiences are now the most offered benefit, at 62% [14].

None of these are free, though. Lounge passes and service perks still cost money, so model them like any other benefit. But many of them cost far less than the points they replace, and members feel them without having to calculate anything. That's the same reason being recognized tends to stick with people longer than another nudge does.

Tell members early, and tell them plainly

In my experience, members will forgive a lot if they understand what's changing and why, so the way you announce it deserves as much planning as the change itself.

• Give real notice. Starbucks gave members about seven weeks before it raised the stars needed for rewards in 2023, but the notice was a short email about updated terms, and members voiced their displeasure [15]. Notice works best when it actually explains the change. Expedia gave a little over a month [2]. For me, coming from tech and retail, I'd aim for 60 to 90 days at a minimum. Adjust that based on your vertical.
• Lead with the new benefits. Launch them before or alongside the change, so members see what they're getting at the same moment they learn what's changing.
• Explain it in plain language. Say what's changing, when, and why.
• Apologize where it's warranted. In the demotion research, an apology reduced the negative reaction where compensation had no effect [6].
• Show each member their path. Tell people exactly what it takes to keep their benefits or their tier. If you're redesigning tiers, my post on building a tier ladder covers how to keep the steps reachable.

Try the new mix on one segment first

A reward structure change is a program change, so test it like one. Give the new benefit mix to one segment, keep a holdout on the current program, and watch purchase frequency, redemption and complaints over at least one full purchase cycle. I walked through the setup in this post on measuring program lift, so I’ll spare you recapping that here.

Not every member wants softer perks

Antavo's 2026 research found that saving money is still the main reason people join a program [12]. For members who joined for the discount, early access may not feel like much, so look at your best customers, who need the least, separately from the members who are still deciding how they feel about you, before you decide what to change.

If a lot of your sales come through retailers, look at where your members earn. A member scanning receipts from the grocery aisle and a member buying on your site might feel the same change very differently, so look at and evaluate each group before you touch anything.

Now come back to the point I made at the top. Most of the public examples are airlines, hotels and restaurant chains, where members complain loudly. In a consumer brand's program, the drop usually shows up as members buying less often, which looks a lot like the members I wrote about who sign up and then go quiet. That's why the holdout in the last section matters so much, because without it you'd have nothing to compare against.

The order I'd make the changes in

If I were planning a reward change for 2027, I'd get the new benefits locked and ready first, so the announcement leads with something real that members are getting. Next comes the change to how members earn going forward, while the points and status they've already built up stay where they are. Members should get 60 to 90 days of notice, but check your program terms first, since some set their own notice period, and you should be able to explain the change in one plain sentence before it goes out. If you can’t, re-evaluate your messaging for simplicity, for your members and for your own sanity.

Look closely at which members will feel the change the most. Make sure they get something back, or feel some extension of value from the brand ahead of the rollout.

And before any of it reaches the whole base, run the test. I’m sure you’ll learn something that helps you fine-tune the plan.

If you'd like someone to pressure-test your reward changes before members see them, our team can walk through it with you.

Frequently Asked Questions

How do you decide which loyalty program benefits to cut?

Start with earn on low-margin categories, rewards almost nobody redeems, and blanket discounts that mostly go to members who would have bought anyway. Protect points and status members have already earned, and test the change on one segment with a holdout before rolling it out.

What is a loyalty program devaluation?

A devaluation is any change that makes a program's rewards less valuable to members, such as lowering earn rates, raising the points needed for a reward, or making status harder to reach or keep.

How do you reduce loyalty program costs without losing members?

Cut earn on low-margin categories and rewards nobody uses, protect what members have already earned, add low-cost benefits like early access and service perks, give 60 to 90 days of notice, and test the change with a holdout first.

How much notice should members get before a loyalty program change?

There's no single rule for consumer brands, but 60 to 90 days is a reasonable minimum. Notice alone won't prevent a backlash, so pair it with a plain explanation and any new benefits launching at the same time.

Why is losing loyalty status so damaging?

Research on status demotion found the negative effect of losing status is stronger than the positive effect of gaining it. In a field study with a European department store chain, demoted customers spent 17.9% less and made 21.5% fewer transactions than in the same period the year before.

Which loyalty program perks cost the least to offer?

Early access to launches, restocks and sales, service perks like longer return windows or priority support, automatic protections like price adjustments, and more time to use earned rewards. Many of these perks cost far less than the points they replace, though none of them are free.

Sources

1. Gartner, 2026 CMO Spend Survey, May 11, 2026.
2. AwardWallet, June 26, 2026.
3. View from the Wing, Oct 6, 2026.
4. CX Dive, Oct 7, 2026.
5. Traveling for Miles, Oct 9, 2026.
6. Wagner, Hennig-Thurau and Rudolph, "Does Customer Demotion Jeopardize Loyalty?" Journal of Marketing, May 2009.
7. Fortune, Oct 18, 2023.
8. Business Traveller, Sept 28, 2023.
9. Goodwin, on the CFPB circular of Dec 18, 2024, January 2025.
10. Shelper, Loyalty Programs: The Complete Guide, 3rd edition, Chapter 1.
11. Bond Brand Loyalty, June 3, 2026.
12. Antavo, Global Customer Loyalty Report 2026.
13. Restaurant Dive, April 13, 2026.
14. EY, 2026 Loyalty Market Study, p. 21.
15. Boston.com, Jan 14, 2023.