On a Reddit thread in late 2023, a Greggs customer stood at the counter with a valid birthday reward in the app and could not work out how to redeem it. Neither could the staff member serving them. The reward expired. That single interaction, buried in a niche subreddit, tells you more about running an app-based loyalty programme than any feature list Greggs has ever published.
Greggs Rewards is a six-category stamp programme (Hot Drinks, Breakfast, Pizza, Sweet Treats, Sandwiches and Other), delivered through a proprietary iOS and Android app with over 15 million downloads. Every seventh purchase in a category earns a free item. It works at Greggs' scale because a dedicated CRM function, in-house app team and 2,500-plus shops absorb the complexity. Below is what independent hospitality operators can actually take from it, and what will silently cost you money if you try to copy the mechanic without the machine behind it.
- Greggs' six-category stamp card is a scale mechanic, not a small-group mechanic. It fragments purchase behaviour deliberately to collect category-level data.
- The programme is not the app. It's a full-time product with a marketing team, ops training and continuous content behind it.
- For groups of 5 to 50 sites, a simpler stamp or points structure will out-perform a Greggs clone on repeat rate and staff execution.
- The biggest operational risk isn't the tech. It's front-of-house staff not knowing how to redeem, which kills trust faster than any bad UX.
- Loyalty data is only valuable if someone is paid to act on it weekly. Otherwise you're collecting exhaust.
How Greggs Rewards Actually Works#
Strip the branding away and Greggs Rewards is a category-segmented digital stamp card. Buy six items in a category and the seventh is free. The categories, per the Greggs app page, are Hot Drinks, Breakfast, Pizza and Sausage Rolls, Sweet Treats, Sandwiches and Salads, and Other. There's also a birthday reward, occasional app-exclusive offers and mobile order-ahead.
The mechanic is deliberately fragmented. A customer buying a coffee every morning and a sausage roll twice a week is filling two separate cards at different speeds. To Greggs, that produces clean category-level purchase data across millions of transactions. To the customer, it produces the birthday-reward-at-the-counter problem noted in this Reddit thread: too many small rewards, unclear redemption path, staff not always sure what to do.
What Greggs Got Structurally Right#
Three decisions deserve real credit, and all three are strategic rather than technical.
- Category segmentation as a data play. Greggs isn't trying to make loyalty simple. It's trying to learn which categories drive repeat, at branch level, across a national footprint. The stamp card is the collection instrument.
- App-exclusive gifts and offers. Per Greggs' promotions terms, the free items and birthday rewards are only available in-app. That gates the value behind account creation, which is how you convert anonymous walk-ins into identified, addressable customers.
- The programme is treated as a product. There's a team behind it. Continuous release notes, new campaigns, seasonal offers, order-ahead. It doesn't sit still, which is why it keeps 15 million downloads engaged rather than churning.
"The Greggs programme only works because someone at Greggs is running it every single day. That's the part no competitor teardown mentions."
The Hidden Cost: What Scale Is Actually Paying For#
The feature-list version of this teardown ends at 'six categories, free seventh item, app-exclusive rewards.' The operator version starts there. Because to run what Greggs runs, you need:
| Function | What Greggs has | What most independent groups have |
|---|---|---|
| App development | In-house or dedicated agency, continuous release cycle | None, or one-off build sitting untouched |
| CRM / loyalty ownership | Full-time CRM team, category managers | Marketing manager wearing five hats |
| Content and campaigns | Weekly, tied to menu launches | Ad-hoc, when someone remembers |
| Staff training on redemption | National training programme, POS-integrated prompts | Verbal handover, high turnover, gaps |
| Data interpretation | Analytics team turning category data into merchandising decisions | CSV exports no one opens |
This is the trap. Operators see the Greggs mechanic, buy loyalty software that can replicate it, launch it, and then discover the software was five percent of the actual job. The other ninety-five percent (running it, promoting it, training staff, reading the data, adjusting the rewards) has no owner. Twelve months later the programme has 8,000 sign-ups, a two percent monthly active rate, and no one can prove it drove a single incremental visit. That's the pattern behind how loyalty actually compounds into repeat revenue, or fails to.
The Redemption Problem Independents Should Take Seriously#
Go back to the Reddit customer with the un-redeemable birthday reward. At Greggs, that's a rounding error. Millions of successful transactions absorb it. At an independent group with 12 sites, that same interaction is a churned regular. Front-of-house friction at the moment of redemption is the single most-underestimated failure mode in hospitality loyalty.
The lesson isn't 'don't do app-based loyalty.' It's: the redemption moment must be so simple that a new hire on day two can execute it without hesitation. A wallet pass with a scannable QR, checked once against a staff scanner, beats an in-app modal with codes and expiry windows every time. The stamp card mechanics vs points comparison is really a comparison of redemption complexity.
What Independent Groups Should Actually Build Instead#
Skip the six-category clone. Below is what an independent group of 5 to 50 sites should build if the goal is repeat rate and LTV, not data collection for its own sake.
| Design choice | Greggs approach | Better for independents |
|---|---|---|
| Card structure | Six category stamp cards | One stamp card, or a points card with two reward tiers |
| Delivery | Native iOS / Android app, 15M downloads to maintain | Apple Wallet and Google Wallet pass, no app to download |
| Sign-up friction | App download, account creation, email verification | Short link, web form, pass in wallet in under 60 seconds |
| Rewards | Free 7th item per category | One clear reward customers can describe from memory |
| Staff process | In-app scan, category logic, redemption modal | Staff scanner adds stamp or redeems, one action |
| Data goal | Category-level national trends | Identify top 20% of customers, drive their next visit |
The point isn't that Greggs is wrong. It's that Greggs is solving a different problem. Their loyalty programme is a data instrument feeding a merchandising engine. Yours needs to be a repeat-visit engine feeding your P&L. Different jobs, different designs.
The Operator Trap: Building an App vs. Running a Programme#
Two founders of hospitality groups can buy the same loyalty software on the same day. Twelve months on, one has a programme driving 22% of monthly covers and a clear picture of who their best customers are. The other has a dashboard they log into twice a year. The software was identical. The difference was operational: who owns it, who writes the campaigns, who trains staff, who reads the redemption data on a Monday.
This is why hospitality operators keep buying loyalty tools and keep feeling burned by them. The tool worked. The programme didn't, because no programme runs itself. Greggs knows this. Their competitive advantage isn't the app, it's the team behind the app. For independent groups, the equivalent is having loyalty run for you, not by you, so the daily work actually happens without adding a headcount you can't afford.
How Carrott Delivers the Outcome Without the Greggs-Sized Team#
Carrott is a wallet-native loyalty platform (cards live in Apple Wallet and Google Wallet, no app for customers to download) combined with programme management. We build the mechanic, run the campaigns to the wallet pass, and design the redemption flow so staff can execute without training drift. That's the model. See what we built for independent hospitality groups for the shape of the work.
Not everything Greggs does is possible without a Greggs-sized team, and it shouldn't be your goal. What is possible: a programme customers can join in 60 seconds, redeem without confusion, and that produces enough signal for you to know which visits it's driving. That's the outcome. The six-category stamp card is not.
Is the Greggs app worth copying for an independent bakery or cafe group?
The mechanic (stamp card, free reward) is worth understanding. The delivery (dedicated app, six categories, in-house team) is not worth copying below roughly 500 sites. Below that scale, a simpler wallet-based programme with one clear reward outperforms a Greggs clone on repeat rate and staff execution.
Why does Greggs use six separate stamp cards instead of one?
To collect category-level purchase data at national scale. Greggs isn't optimising for customer simplicity; it's optimising for merchandising intelligence across 2,500-plus shops. For an independent group, the data volume doesn't justify the customer confusion the segmentation creates.
What's the most common failure point in app-based loyalty programmes?
Redemption friction at the counter. Staff who don't know the process, unclear reward states in the app, expired offers customers didn't realise they had. This is documented directly in Greggs customer forums and is amplified in smaller groups where staff turnover is higher and central training is thinner.
How do I know if my current loyalty programme is actually driving repeat revenue?
Compare the visit frequency of enrolled customers against a matched cohort of non-enrolled customers over 90 days. If enrolled customers aren't visiting more, you have a discount programme, not a loyalty programme. You can <a href="/blog//loyalty-programme-audit">find out if your current programme is actually driving repeat visits</a> with an audit.
Do I need a custom app like Greggs to run a good loyalty programme?
No. Custom app development costs six figures upfront and needs continuous investment to stay useful. Apple Wallet and Google Wallet passes deliver the same core mechanics (real-time updates, push notifications, redemption at counter) without the download friction or maintenance burden.