Australia & New Zealand · Independent retail and hospitality

The Loyalty Playbook.

Almost everyone here is already in a loyalty programme. Almost nobody uses most of them. Seven plays for being the one card that gets used, built on what the research actually shows.

7 plays 11 sources 15 minute read Updated August 2026

The state of play

This is the most loyalty-saturated corner of the world.

90%

of Australians belong to at least one loyalty programme, holding an average of 4.3 memberships each.

Statista · Australian Loyalty Association, 2025

4.7m

Airpoints members in New Zealand. The country has roughly 5.3 million people in it.

Air New Zealand, 2024

~half

of enrolled members actively engage with the programmes they have joined. The rest signed up and stopped.

Statista · Australian Loyalty Association, 2025

Two things follow from those numbers, and they point in opposite directions.

The good news is that you do not have to sell the idea. Woolworths, Air New Zealand and the supermarkets have spent two decades and enormous budgets teaching this market that a loyalty card is normal, worth carrying, and worth asking for at the counter. Nobody needs convincing any more.

The bad news is that the wallet is full. Your customer is not deciding whether to join a loyalty programme. They are deciding whether yours is worth the four seconds and the mental space, when they already have four others they have mostly forgotten about.

Why they get abandoned

It is friction. It was always friction.

The comfortable explanation is that people are tired of loyalty programmes. The research says something more specific and more useful: people are not tired of rewards, they are tired of the work required to get them.

70%

never complete a sign-up because registration takes too long.

47%

name expiring points as their single biggest frustration.

And on apps specifically: four in five people who downloaded one said they did it because they had to, not because they wanted to. Roughly seven in ten reported being annoyed, frustrated or angry about it.

Figures from industry surveys. Sources listed at the end, with the academic and industry work marked separately.

The plays

Seven things that actually move the needle.

Make joining one scan, with nothing to install

Every step between "yes I'll join" and "you are joined" costs you people. An app download costs you most of them. A form asking for a name, an email, a password and a date of birth costs you the rest.

Evidence Seven in ten people abandon sign-up because it takes too long. Long or awkward onboarding is the most cited obstacle to joining at all.
Do this: one QR at the counter, one scan, card lands in the phone wallet that is already installed. Ask for a name and one contact detail. Nothing else.

Give people a head start they did not earn

This is the most reliable finding in the whole field and almost nobody uses it. In a field study of a cafe loyalty card, customers given a twelve-stamp card with two stamps already filled in completed it faster than customers given an empty ten-stamp card. Both groups had to buy exactly ten coffees.

Evidence · peer reviewed Kivetz, Urminsky & Zheng (2006), Journal of Marketing Research. Endowed progress accelerated real purchases, not just stated intentions.
Do this: if you want ten purchases, do not build a ten-stamp card. Build a twelve-stamp card and put two stamps on it the moment someone joins.

Never expire anything

Expiry exists to protect a liability on a balance sheet. For a single-site independent, that liability is small and the goodwill you burn is not. A customer who discovers their stamps vanished does not quietly accept it. They tell people.

Evidence Expiring points are the most cited frustration with loyalty programmes, named by roughly half of surveyed consumers.
Do this: no expiry. If you must cap your exposure, cap the reward value rather than deleting what someone has already earned.

Set the finish line where people can see it

The goal-gradient effect only helps you if the goal is visible and believable. A reward at fifteen visits is invisible to someone on visit two. A reward at five is a plan they can hold in their head.

Match the threshold to how often your trade actually sees people. A cafe can ask for eight or nine because a regular gets there in a month. A barber asking for eight is asking for a year and a half of loyalty on faith.

Do this: aim for a first reward inside six to eight weeks for a typical customer. Work backwards from that to your number.

Price the reward between one and three percent

The most common way an independent loses money on loyalty is picking a reward that feels generous without ever working out what it costs as a share of takings. A free coffee every fifth visit is a twenty percent discount wearing a disguise.

One to three percent of what enrolled customers spend is the band where a programme pays for itself. Below one percent it is not motivating. Above three, you are buying visits you would have got anyway.

Do this: reward value divided by (threshold × average spend). If that is over 3%, raise the threshold or shrink the reward.

Put it where they already look

A card in Apple Wallet or Google Wallet is on a phone that is already unlocked forty times a day. It can surface on a lock screen near your door. A paper card is in a drawer and an app is on page four behind a folder called "Stuff".

Evidence Around three in ten abandonments are attributed to inconvenient apps, and four in five app downloads were felt to be compulsory rather than wanted.
Do this: wallet pass, not app. And keep the messages rare enough that they stay welcome.

Notice people leaving before they have gone

A cardboard punch card tells you nothing. You find out someone stopped coming in about four months after they stopped, usually by accident, usually too late to do anything about it.

The whole practical advantage of a digital card is that lapsing is visible while it is still reversible. Someone who came in weekly and has not been seen for three weeks is a specific person you can send a specific reason to return.

Do this: set a lapse window that matches your trade, roughly three times a normal gap between visits, and act on it while they still remember you fondly.

The maths, once

A worked example, so play five is not abstract.

A cafe selling flat whites at A$5.50 wants to know what a buy-nine-get-one-free card costs.

A$5.50

Reward value, one flat white

9

Purchases before the reward

A$49.50

Customer spend to earn it

11.1%

Effective reward rate. Far too high.

Eleven percent is not a loyalty programme, it is a standing discount you are giving to people who were already coming in. The fix is not to make the card meaner. It is to make the reward smaller and more frequent, or the threshold higher and the reward genuinely worth reaching.

The same cafe offering a free coffee at every twentieth visit lands at 5%. Offering a free filter coffee, worth A$4, at twenty visits lands at 3.6%. Adding a small food item instead, worth A$3, at eighteen visits lands near 3%.

Most independents have never done this sum. It takes ninety seconds and it is the difference between a programme that pays and one that quietly bleeds.

Run your own numbers

The bit nobody writes about

Collecting contact details puts you under two sets of rules.

A loyalty programme is a marketing database. The moment you hold a customer's email or mobile and send them anything, you are inside legislation that most independent operators have never read. None of this is difficult, but getting it wrong is expensive and the excuse of not knowing does not work.

Australia

Spam Act 2003

Applies to any commercial electronic message with an Australian link. Three obligations, all of them absolute:

  • Consent, either express or reasonably inferred. Record when and how it was given.
  • Identify yourself clearly, including accurate contact details.
  • A working unsubscribe that stays working for at least 30 days.
New Zealand

Unsolicited Electronic Messages Act 2007

The same shape as the Australian rules, and it applies to messages sent from, or to, New Zealand. Consent, identification, and a functional unsubscribe.

If you trade on both sides of the Tasman you are subject to both, so build to the stricter reading and stop worrying about which applies.

Both

Privacy Act 1988 (AU) & Privacy Act 2020 (NZ)

Customer names, contact details and visit histories are personal information. You need to be able to say what you hold, why, and delete it on request.

  • Only collect what the programme actually needs.
  • Be able to hand someone their own data if they ask.
  • Do not sell it, and do not share one venue's list with another.

This is a plain-language summary to help you ask the right questions, not legal advice. If you are running marketing at any scale, have someone qualified look at your consent flow once.

Sources

Where every number came from.

Peer-reviewed work and industry surveys are not the same thing and we have not blended them. Market-research figures are marked so you can weigh them accordingly.

  • 01Peer reviewed
    Kivetz, R., Urminsky, O. & Zheng, Y. (2006). The Goal-Gradient Hypothesis Resurrected: Purchase Acceleration, Illusionary Goal Progress, and Customer Retention. Journal of Marketing Research, 43(1), 39–58. Source for plays two and four.
  • 02Market research
    Statista, loyalty programme enrolment and memberships per person, Australia, 2016–2025. Source for the 90% and 4.3 figures.
  • 03Market research
    Australian Loyalty Association, Australia Loyalty Programs Market Report 2025. Source for active engagement running near half of enrolment.
  • 04Company reported
    Air New Zealand, Airpoints membership, 2024. Source for 4.7 million members.
  • 05Company reported
    Woolworths Group, Everyday Rewards New Zealand active membership, 2024. 1.8 million active members.
  • 06Industry survey
    Reported consumer research on registration abandonment, expiring points, onboarding difficulty and app inconvenience. Sources for plays one, three and six. These are vendor and trade-press surveys rather than academic work, and methodology is not always published.
  • 07Industry survey
    Reported research on compulsory app downloads and user sentiment. Source for the four-in-five figure in play six.
  • 08Legislation
    Spam Act 2003 (Cth), Australia.
  • 09Legislation
    Unsolicited Electronic Messages Act 2007, New Zealand.
  • 10Legislation
    Privacy Act 1988 (Cth) and the Australian Privacy Principles.
  • 11Legislation
    Privacy Act 2020, New Zealand.
What is missing None of this measures independent venues specifically. Every consumer figure above describes loyalty in general, which in this market mostly means supermarkets and airlines. Nobody has published research on what actually happens at a forty-seat cafe or a two-chair barber, so we have built the plays on the closest evidence available and told you which is which. Where a play rests on a trade survey rather than a controlled study, it says so.

Put play five through the calculator

Seven plays, or one card that does all of them.

One scan to join, nothing expires, it lives in the wallet they already carry, and it tells you who stopped coming in. Free for sixty days.

Build my card Run the numbers