7 Types of Loyalty Programs (And How to Choose the Right One)

Every loyalty program is designed to change customer behaviour: the problem is that the behaviour worth changing isn't the same for every business.

A brand chasing purchase frequency needs a different model than one trying to stand out in a crowded, low-frequency category. A B2B manufacturer selling through distributors needs something entirely different from a bank trying to drive app adoption. 

A technically flawless program can still fail if its core mechanic isn't aligned with the behaviour you're trying to change.

This guide breaks down the seven program types that cover almost every loyalty model in the market today, shows how real brands have used each one, and gives you a framework for working out which one actually fits your business. 

If you're earlier in the process and want the full picture of how loyalty programs work before picking a type, our Loyalty 101 guide is the place to start. If you already know your type and want to build out the strategy around it, jump to our definitive guide to loyalty strategy once you're done here.

Loyalty program types at a glance

Earn & burn (points-based) loyalty programs

An earn & burn program rewards customers with points, cashback, or credit for every qualifying purchase, which they then redeem for a reward once they hit a threshold. It's the oldest and still the most common loyalty model, mostly because it's easy for customers to understand and easy for brands to launch quickly.

Earn & burn is strongest in categories where customers already purchase frequently. In low-frequency categories, customers may take too long to earn a meaningful reward, which makes the program harder to sustain.

ARDEX and BAL, two building products brands under the ARDEX Group, built GivBax Rewards on this model to reach end customers who normally buy through distributors. Using receipt scanning to verify purchases from any UK distributor, the program paid out cash rather than points, redeemable at over 18,000 ATMs. 

Within six months it had a 73% engagement rate, 17% of the addressable market signed up, and total spend per user rose more than 10%. 

It's a good reminder that earn & burn isn't only a consumer retail play: it works for B2B too, once you solve the harder problem of reaching end customers who buy through someone else.

AkzoNobel took a lighter version of the same approach with its Dulux Cashback campaign, built and launched in under 10 days. Customers scanned receipts from any UK retailer to claim cashback, and the campaign pulled in 3,500+ site visits and 600+ registrations without requiring a long-term loyalty infrastructure build.

Watch out for: setting your earn-to-redeem ratio without testing it first. Too stingy and members disengage before their first reward. Too generous and the program becomes a discount scheme that erodes margin.

Tiered loyalty programs

Tiered programs sort customers into levels with better benefits unlocked at each level. The appeal is pretty straightforward: customers have a reason to spend more to reach the next tier, then another reason to stay there.

SKB Bank built tiering into its BONUS Rewards pilot, where new customers had to complete three tasks to move from a Basic tier up to Bronze before unlocking further rewards. SKB is also a good example of why loyalty mechanics rarely operate in isolation: the company layered tiers on top of gamified challenges, which is closer to how most real tiered programs actually work in practice (more on combining types further down).

This model earns its keep in categories where status has real social currency: travel, hospitality, premium retail, financial services. It's less effective in commodity categories where customers don't feel a status difference between tiers, only a rewards difference, in which case a simpler earn & burn model usually performs just as well for less complexity.

Watch out for: launching tiers before you have enough purchase history to set thresholds sensibly. Get the thresholds wrong and you'll either have almost everyone stuck at the bottom tier or almost everyone maxed out at the top within weeks, and both outcomes kill the aspiration mechanic that makes tiering work.

Perks and paid membership programs

Perks-based programs, sometimes called paid or VIP membership, ask customers to pay a recurring fee in exchange for a fixed set of ongoing benefits: free shipping, exclusive pricing, early access, or premium service. Amazon Prime is the most recognisable example, but the model shows up across retail, media, and increasingly B2B too.

Customers who pay for membership tend to use it more, spend more, and churn less than customers in free programs, because the upfront cost creates a sense of ownership over getting value back out. But the model only works when customers can see a clear return on the membership fee. If members do the maths and conclude the fee barely pays for itself, the program becomes a churn risk rather than a retention tool.

This model tends to suit high-frequency categories and subscription-adjacent businesses best, where customers are already used to paying recurring fees and can quickly rack up enough usage to justify the cost. It's a harder sell for infrequent or big-ticket purchase categories, where customers won't interact with the program often enough to feel the value.

Watch out for: launching a paid tier without a free or low-cost entry point first. Asking for money before customers have any track record with your brand is a much harder conversion than asking once they're already engaged.

Gamified loyalty programs

Gamified programs use game mechanics, challenges, badges, streaks, and levels, to drive engagement beyond the purchase itself. The reward isn't always a discount: sometimes it's simply unlocking the next challenge or climbing a leaderboard, which is part of what makes this model useful for driving behaviours that don't have an obvious monetary reward attached.

Biscuit, a pet wellness app, built its entire loyalty layer around gamification. Members join walking challenges, complete quizzes, and collect badges to earn XP, unlocking new levels and rewards along the way. In nine months, 

Biscuit reached over 50,000 users, who between them walked more than 950,000 miles and completed over 60,000 weekly challenges, all while maintaining a 4.6-star app rating. Biscuit shows that gamification can drive engagement without relying on a traditional points-for-purchase system.

SKB Bank's BONUS Rewards program applied a similar approach to a very different category: banking.

By turning transactional and non-transactional actions (card usage, referrals, GDPR consent) into completable challenges, SKB saw Flik payment transactions grow 188% and app activations rise 17%, with 75% of participants reporting satisfaction with the program specifically because of its game-like feel.

Watch out for: gamifying a category where the underlying frequency is too low to sustain the mechanic. Challenges and streaks need regular touchpoints to work. If customers only interact with your brand twice a year, there's no streak to protect.

Coalition loyalty programs

A coalition program lets customers earn and redeem one shared currency across multiple, often unrelated, brands. Instead of collecting separate points with every retailer, the customer earns faster because more of their everyday spending counts, and participating brands benefit from shared infrastructure and a bigger combined rewards catalogue than any one of them could offer alone.

Tickit, built for Dubai Holding, is one of the clearer examples of this model working at scale. Using card-linking technology with Visa and Mastercard, members earn rewards automatically just by shopping at any of more than 3,000 participating outlets across 300 brands and 10 sectors in the UAE, with no receipt scanning or app tap-in required. 

Within a year of launch, Tickit became the second-largest loyalty program in the UAE by outlet count.

Coalition mechanics don't have to operate at national scale: Wotzon Rewards applies coalition mechanics to tourism, giving visitors money-saving rewards across a network of local businesses that individually would struggle to build or justify their own loyalty infrastructure.

Watch out for: underestimating the operational complexity. Coalition programs need every partner brand aligned on redemption value, funding splits, and data sharing, and that alignment is usually harder to maintain than the technology itself.

Referral and advocacy programs

Referral programs reward existing customers for bringing in new ones, usually with a reward for both parties once the referred customer completes a qualifying action. 

Referral is often less a standalone loyalty program than a growth mechanic layered onto an existing one.

SKB Bank's referral mechanic, run alongside its gamified tiers, got 3.5% of program participants to refer a friend, converting into 1.48% new-to-bank clients overall. That's a meaningful acquisition channel for a bank, where new customer acquisition costs are typically high and trust matters more than most other categories.

Referral works best when the product or service is something customers want to recommend anyway. It works poorly when it's used to paper over a program that isn't otherwise compelling, since asking customers to vouch for something they're lukewarm on rarely goes well.

Watch out for: rewarding the referral action itself rather than the outcome. Paying out before the referred customer actually engages invites fraud and low-quality signups that don't convert into real value.

Hybrid and omnichannel loyalty programs

Most mature loyalty programs, including several already mentioned in this guide, aren't a single type at all. 

They combine two or three mechanics: points as the foundation, tiers for aspiration, gamified challenges for engagement, referral for growth. Hybrid programs combine multiple mechanics because customer behaviour rarely fits a single model.

Ravensburger shows a different kind of hybrid: not multiple mechanics, but one mechanic (points-for-purchase) scaled consistently across markets and channels. The toy and puzzle manufacturer relaunched its loyalty program across 11 European and US markets, connecting online and in-store purchasing into a single view and localising currency, language, and reward catalogues for each region. 

In its first year, the relaunched program drew 127,000+ registrations, a 58% engagement rate, and a 20% repeat purchase rate, strong enough that Ravensburger is now planning a further rollout into 20 more countries.

Watch out for: adding complexity for its own sake. A hybrid program should combine mechanics because your customer behaviour genuinely calls for it, not because more features look more impressive in a pitch deck.

How to choose the right type for your business

The best model depends on three variables: your business model, purchase frequency, and the behaviour you want to change.

• By business model. B2B and trade businesses selling through distributors, like ARDEX and BAL, often need earn & burn or spend-based mechanics that reach end customers indirectly. Multi-brand groups, malls, and franchises are natural fits for coalition. Single-brand consumer businesses have the widest range of options open to them.
• By purchase frequency. High-frequency categories (grocery, QSR, FMCG) can sustain gamification and tiering because customers interact often enough for streaks and status to mean something. Low-frequency, high-value categories (furniture, cars, home renovation) usually do better with simpler earn & burn or spend-based models, since anything that depends on regular touchpoints will struggle for lack of them.
• By primary goal. If the goal is first-party data capture, earn & burn with receipt scanning or card linking, like PepsiCo's approach, is hard to beat. If it's transaction volume and habit formation, gamification has the track record (Biscuit, SKB Bank). If it's lower-cost acquisition, referral mechanics deserve a bigger role than they usually get. If it's building a defensible, everyday-relevant network, coalition is worth the operational overhead.

Can you combine loyalty program types?

Yes, and in practice most successful enterprise programs already do. SKB Bank combined tiering, gamification, and referral in a single pilot. Ravensburger combined a simple earn & burn mechanic with full omnichannel and multi-market complexity. 

Neither started that way. Both began with a core mechanic and added others once the first one proved out.

That's usually the right order of operations: get one mechanic working and measurable before layering on a second. Combining types from day one, before you know what your customers actually respond to, tends to produce a program that's expensive to run and hard to diagnose when something isn't working.

Common mistakes when choosing a loyalty program type

Even a well-designed program can underperform if the mechanic doesn't fit how your customers actually buy and engage. Here are some of the most common mistakes to avoid.

• Copying a competitor's model without checking the fit. A tiered program that works brilliantly for an airline won't automatically work for a mid-frequency retailer. Match the mechanic to your own purchase frequency and margin profile, not someone else's.
• Over-gamifying a low-frequency category. Challenges and streaks depend on regular interaction. Force them onto a category where customers only show up twice a year and you'll get a program nobody finishes.
• Setting tier thresholds before you have the data. Guessing at spend thresholds without purchase history usually produces one of two bad outcomes: almost nobody reaches the top tier, or almost everyone does within the first quarter.
• Assuming B2B customers don't want loyalty programs. ARDEX and BAL's results say otherwise. B2B buyers respond to recognition and reward just as consumer customers do, provided the mechanic accounts for how they actually buy.