Key Takeaways
Americans join readily and disengage quietly. Every 2026 survey agrees that roughly half or fewer of their memberships are actively used.
Unredeemed points are not a saving. They are a balance-sheet liability and a signal that members cannot see a reward worth reaching.
US shoppers rank practical value such as free or faster shipping above aspirational rewards. Lead with the friction you remove.
Time to first redemption is the most actionable number in the programme. Shorten it and enrolments become members.
AI earns its place in next-best-offer, churn prediction, fraud detection and liability forecasting. Earn rates and rules stay with humans.
Every trend here depends on identity resolution. If a member looks like three records, none of it works.
US loyalty is in an unusual position. Enrolment has never been higher, consumer appetite for joining programmes is rising, and marketers are putting a large share of budget behind them. Active participation, meanwhile, has barely moved. Most Americans belong to far more programmes than they actually use.
That gap is the story of the sector right now, and it is what every trend below is trying to close. The programmes gaining ground are not the ones with the richest earn rates. They are the ones that make value easy to see, easy to reach and easy to take.
This guide covers the state of US loyalty, what customers say they want, the ten mechanics that are changing, where AI genuinely helps, and what to do about it. For the programme fundamentals, see our guide to customer loyalty programs.
The state of US loyalty in 2026
Three findings describe the market more usefully than any single statistic. They come from different surveys with different samples, which is what makes the pattern worth trusting.
Enrolment is up, active use is not
Industry surveys disagree on the exact number of programmes the average American belongs to, which is itself informative. Estimates in circulation for 2026 range from about eight to more than seventeen, and every one of those sources agrees that roughly half or fewer are actively used. The spread comes from different sampling and different definitions of membership, so treat any single figure with caution. The direction is consistent: people join readily and disengage quietly.
Antavo's 2026 global customer loyalty report, which includes a US sample, found that 43.2% of consumers said they were more likely to join a loyalty programme than in the previous year. Bond's US member research has long reported that a large majority say a programme makes them more likely to keep doing business with a brand, and that a smaller but substantial share say they spend more because of it. These are vendor surveys with the self-selection that implies, but the pattern across them is stable.
Where the unredeemed value sits
Unspent points are the clearest measure of the engagement gap. Antavo's 2026 reporting puts the value of unspent loyalty points in the billions of dollars annually, and separate estimates place the share of points earned but never redeemed at roughly a quarter. For a programme owner this is not good news disguised as savings. Unredeemed points are a liability on the balance sheet and a signal that members cannot see a reward worth reaching for.
What marketers are spending
Statista reports that US marketers dedicate close to 60% of marketing budget to loyalty initiatives, and projects the US loyalty market to grow at a compound annual rate in the mid-to-high single digits through the second half of the decade. Antavo's self-reported ROI figure for loyalty programmes sits above five times, broadly flat year on year. The money is not the constraint. Attention is.
~50%
Of memberships actively used
consistent across every 2026 survey
43.2%
More likely to join than last year
Antavo 2026 report
~1 in 4
Points earned but never redeemed
the engagement gap, in liability terms
What US customers want from a loyalty program
Consumer research in this market is unusually consistent about priorities, and the order is not the one most programmes assume. Practical value sits above aspirational value in nearly every study that asks.
Practical value before aspirational value
Attentive's 2026 consumer research puts free or faster shipping at the top of what US shoppers want from a programme, ahead of rewards points. That ordering matters. The thing customers rank first is a removal of friction they experience on every order, not a prize they might reach eventually. Programmes that lead with aspiration and bury the practical benefit are optimising for the wrong half of the base.
Flexibility in how value is taken
The strongest preference signal across US research is choice at redemption: statement credit, gift cards, donation, partner transfer, or spending points as part-payment at checkout. A single-currency catalogue with one redemption route reads as restrictive to a customer who has been trained by every other programme they belong to.
Recognition that is not a discount
Status, early access, priority service and experiences carry perceived value well above their cost, and they do not train price sensitivity the way a discount does. This is the main argument for building tiers rather than deepening earn rates. Our guide to tiered loyalty programs covers the structure.
Loyalty behaviour by generation
| Group | What tends to drive participation | Design implication |
|---|---|---|
| Gen Z | Mobile-native, social proof, instant and visible value | Micro-rewards, in-app earning, no paperwork |
| Millennials | Experiences, personalisation, flexible redemption | Partner options and experiential rewards at tier |
| Gen X | Practical savings, reliability, multi-brand usefulness | Clear cash-equivalent value and no expiry surprises |
| Boomers | Simplicity, trust, familiar redemption routes | A legible base mechanic and a non-app route to join |
The practical reading of that table is not to build four programmes. It is to make one programme legible at the base and rich at the edges, so the simplest member can use it without help and the most engaged one still finds depth.
Close the gap between earning and using
Broad earn and burn, part-payment at checkout, pooling and micro-rewards, on one platform.
Book a demoTrusted by 5,000+ companies
The ten trends shaping US loyalty programs
These are mechanics rather than themes, which means each one is a decision a programme owner can actually make this quarter. Almost all of them point in the same direction, which is shortening the distance between earning something and using it.
1. Mobile-first participation
The programme now lives on a phone or it does not live anywhere. That means wallet passes rather than cards, push rather than email for time-sensitive value, and an earning flow that survives a bad signal at a checkout counter. A programme whose core journey requires a desktop session has already lost the majority of its base.
2. Embedded loyalty
Earning and redeeming inside the transaction rather than alongside it: points applied at checkout, rewards surfaced inside the banking or retail app, benefits that apply automatically without a code. Every step a member has to take separately is a step where the programme loses them.
3. Instant gratification
Time to first reward is the single most actionable number in a loyalty programme. The longer the first redemption takes, the smaller the share of enrolled members who ever reach it, and the larger the unredeemed liability. US programmes are compressing that distance deliberately, often with a small welcome reward that costs little and converts an enrolment into a user.
4. Flexible earn and burn
Members increasingly expect to earn through more than purchases, and to burn through more than a catalogue. Earning on reviews, referrals, profile completion and app actions, and burning as part-payment, cash back, charity or partner transfer. Flexibility raises redemption rate, which lowers liability and raises engagement at the same time.
5. Interoperable partnerships
Coalition and partner earning make a currency useful outside your own category, which matters most for brands with low natural purchase frequency. The trade-off is commercial and data complexity. Our guide to coalition loyalty programs sets out what the partner route actually involves.
6. Micro-rewards and small redemptions
Low-denomination redemptions, surprise rewards and small unlocks keep the loop running between big redemptions. They are cheap, they are remembered out of proportion to their cost, and they keep members from parking points indefinitely.
7. Points as currency
Points used at checkout as part-payment rather than exchanged for a predefined item. This is the mechanic that makes a loyalty balance feel like money, and it is the clearest answer to the complaint that points are hard to use. It also moves liability off the balance sheet faster, which finance will appreciate.
8. Points pooling
Households and groups combining balances to reach a reward none of them would reach alone. Pooling raises redemption rate and brings in members who had given up on their own balance. It needs fraud controls, because pooling is also a route to transferring value between unrelated accounts.
9. Non-expiring points
Expiry is the most resented mechanic in loyalty and one of the most common reasons members abandon a programme. More US programmes are moving to activity-based expiry, where points persist as long as the member is active, rather than hard annual expiry. Several US states also regulate expiry and dormancy on stored value, so the policy question is legal as well as commercial. Our guide to loyalty program regulations covers that ground.
10. Tiered recognition
Tiers remain the most durable structure in the sector because they reward depth rather than transactions, and because the benefits that matter at the top are operational rather than monetary. The failure mode is thresholds set so high that the mass of the base can never see themselves qualifying.
AI in US loyalty programs
Industry surveys report that around half of loyalty marketers now use AI somewhere in programme management, roughly a third more than the year before. The adoption number is less useful than knowing where it actually pays.
Where AI earns its place
- Next-best-offer selection: Choosing which of a set of approved offers to show a given member. Measurable, reversible and low risk.
- Churn and lapse prediction: Flagging members whose behaviour has changed before they leave, so a retention action has time to work.
- Fraud and anomaly detection: Pattern detection at a scale rules alone cannot cover. Our guide to loyalty program fraud covers the patterns worth watching.
- Segmentation at scale: Behavioural clustering that a human analyst would not have time to maintain weekly.
- Liability forecasting: Breakage and redemption projection, which improves materially with better modelling.
Where it does not
- Setting earn rates: These are commercial and compliance decisions with a direct cost line. They need a human owner.
- Writing programme rules: A rule the team cannot explain is a rule they cannot defend to a regulator or a complaining member.
- Anything without clean data underneath it: A model on fragmented identity data produces confident nonsense faster than a human would.
Our guide to AI in loyalty programs goes into the implementation detail.
New entrants changing the competitive set
Loyalty is no longer only a retail and travel concern. Three groups have moved in, and all three raise the baseline expectation your programme is judged against.
- Fintech and neobanks: Rewards built into the payment itself, with no separate enrolment step, which has normalised embedded earning.
- Marketplaces and quick commerce: Subscription benefits that combine shipping, pricing and content, which has normalised paying for membership.
- Brands with no purchase frequency: Insurers, utilities and healthcare entering loyalty to create contact where there was none, usually through behaviour-based earning rather than spend.
How to act on these trends
None of the trends above works on fragmented data or an unresolved privacy position. These four rules are the preconditions rather than the programme itself.
Fix data integrity before anything else
Every trend above depends on recognising the same person across channels. If a member looks like three separate records, personalisation misfires, pooling is unsafe, and liability reporting is wrong. Identity resolution is the unglamorous precondition for all of it.
Make the privacy position explicit
US state privacy laws now give consumers access, deletion and opt-out rights that apply directly to loyalty data, and several states regulate what a programme may charge or deny when a consumer exercises them. Settle consent, retention and the partner data-sharing position before building personalisation on top of them.
Remove the silos that split the member
Loyalty data that lives apart from CRM, ecommerce and service data produces a programme that contradicts the rest of the business. The integration work is usually what separates a programme that runs campaigns from one that changes behaviour.
Leave room to learn
Run a holdout group as standard, change one mechanic at a time, and keep the base rule stable while you experiment at the edges. Devaluations and rule changes without notice are the most complained-about events in loyalty, so make the experimentation visible only where it adds value, never where it takes value away.
Metrics to watch
| Metric | Why it matters in 2026 |
|---|---|
| Active member share | The real number behind enrolment, and the gap every trend is trying to close |
| Time to first redemption | The strongest predictor of whether an enrolment becomes a user |
| Redemption rate | Low redemption is not a saving, it is a warning |
| Points liability and breakage | Unredeemed value is a balance-sheet item and an engagement signal |
| Incremental revenue vs control | The only defensible proof that the programme caused anything |
| Mobile share of programme activity | Whether the programme lives where the member does |
| Earn and burn breadth | How many routes members actually use, which drives redemption |
US loyalty on Loyalife
01 · Flexibility
Broad earn and burn
Earning beyond purchases and redemption beyond a catalogue, including part-payment at checkout, cash back, donation and partner transfer.
02 · Mobile
Wallet passes and in-app earning
A programme that works on the phone the member already has, with no desktop step in the core journey.
03 · Structure
Tiers, pooling and micro-rewards
Recognition that rewards depth, household pooling with fraud controls, and small unlocks that keep the loop running.
04 · Intelligence
AI where it pays
Next-best-offer, churn prediction, anomaly detection and liability forecasting, with earn rates and rules under human control.
05 · Data
One identity across channels
Identity resolution and prebuilt integrations so the member is one person to the programme and the business.
06 · Compliance
US privacy and expiry controls
Consent management, state-level rights handling, activity-based expiry and versioned disclosure.
You can also explore loyalty programs and the Loyalife platform.
The through-line in every one of these trends is the same: reduce the distance between earning something and using it. Programmes that shorten that distance convert enrolments into members. Programmes that lengthen it accumulate liability and call it engagement.
Book your Loyalife demo
Frequently asked questions
Estimates in circulation for 2026 range from roughly eight to more than seventeen, depending on the survey's sampling and its definition of membership. The disagreement is worth noting rather than resolving. What every source agrees on is that only about half or fewer of those memberships are actively used, which is the gap the sector is trying to close.
Practical value before aspirational value. Attentive's 2026 consumer research puts free or faster shipping ahead of rewards points, which tells you customers rank the removal of everyday friction above a prize they might eventually reach. After that comes flexibility at redemption and recognition that is not simply a discount.
Usually because the first reward is too far away, the redemption routes are too narrow, or the member cannot easily see what their balance is worth. Antavo's 2026 reporting puts unspent US loyalty points in the billions of dollars annually, with separate estimates placing roughly a quarter of earned points as never redeemed. For the programme owner that is liability, not saving.
Earning through more than purchases, such as reviews, referrals, profile completion and app actions, and redeeming through more than a fixed catalogue, such as part-payment at checkout, cash back, charitable donation or partner transfer. It raises redemption rate, which lowers liability and raises engagement at the same time.
Allowing households or groups to combine balances so they can reach a reward that none of them would reach individually. It raises redemption rate and re-engages members who had written off their own balance. It needs fraud controls, because pooling is also a mechanism for moving value between accounts that have no real relationship.
Hard annual expiry is the most resented mechanic in loyalty and a common reason members abandon a programme. More US programmes now use activity-based expiry, where points persist while the member remains active. Several US states also regulate expiry and dormancy on stored value, so the question is a legal one as well as a commercial one.
Next-best-offer selection, churn and lapse prediction, fraud and anomaly detection, behavioural segmentation at a scale humans cannot maintain, and liability forecasting. Where it does not help is setting earn rates or writing programme rules, both of which are commercial and compliance decisions that need a human owner, and anywhere the underlying identity data is fragmented.
Most widely quoted figures come from vendor-run surveys, including the ROI numbers, and they are usually self-reported by programme operators. The numbers are directionally useful and the cross-survey patterns are stable, but any single figure should be checked against the original report rather than an aggregator blog before you put it in a business case.
Active member share rather than enrolment, time to first redemption, redemption rate, points liability and breakage, incremental revenue against a holdout control, mobile share of programme activity, and how many distinct earn and burn routes members actually use. The control group is what makes any of the rest defensible.
![Diversity and Inclusion in the Workplace: Why It Drives Success [Activities + Survey Tips]](/Blog-cover-light/inclusion-and-diversity-in-the-workplace.avif?dpl=dpl_8fJikYxuCmYv3RQT1yaE4Git6nzR)
![How to Run an Effective Employee Commitment Survey [+ Sample Questions]](/Blog-cover-light/employee-commitment-survey.avif?dpl=dpl_8fJikYxuCmYv3RQT1yaE4Git6nzR)
