How to get customers to join at the counter

Bank Negara Malaysia's annual report for 2025 records that "DuitNow QR transaction volume doubled to three billion in 2025 (2024: 1.5 billion)" and that "there are almost three million registered DuitNow QR touchpoints across Malaysia", adding: "A majority are small businesses". Scanning a code needs no teaching.

The figure includes transactions made by individuals as well as by merchants, and what it measures is paying. It cannot tell you how many customers will scan a second code to join something. It does show that scanning to pay is ordinary now, and a customer who pays that way has the phone out at your counter.

That is the case for putting the join sign beside the payment QR. It is an argument, and no published test of it was found. Two codes side by side bring two problems. One is the wrong scan, so the sign must say what it is in words that can be read from the queue. The other is the wrong scanner. A payment code is scanned inside a banking or e-wallet app. Your join code is a web address and wants the phone's camera, and a banking app's scanner may not open it. Tell customers "with your camera".

In native.cards the scan opens your Business page. In order, the customer meets an email address, your questions, a tick or two, a code and the card.

Joining is by email address. There is no joining by phone number or SMS, so "What's your number?" is the wrong opening at this counter. The customer must be able to open that email on the phone in their hand, because the code is sent to it.

The questions are yours. In program setup, the step headed What do you ask when they join? says: "Every extra question loses a few people." Email is always asked. Their name is ticked as the form opens, and the customer can leave it blank. You can add Phone ("Only if you need to ring or message them."), Existing membership number, Vehicle registration and questions of your own. With the name and all three of those ticked, a customer meets five fields before the code.

A question can also become compulsory without your choosing it. If your published card shows a detail the member has to type, such as their name, the form requires it. A name on the card is a pleasant touch. At the counter it is a second field.

The nearest research comes from online shopping. Baymard Institute, a firm that tests the checkouts of large online retailers, reported in 2024 that the average checkout had 5.1 steps and 11.3 form fields, where "most sites need only 8 form fields in total". Its conclusion: "the number of form fields in a checkout impacts overall usability far more than the number of steps." Those are online checkouts, measured by a firm that sells access to its research, and the article gives no sample size. A counter is plausibly the harder case: the customer is standing, one hand is busy and someone is waiting behind. That step is reasoning. Baymard did not test counters.

If you want a phone number so that staff can find a member who has left their phone in the car, collect it later. The counter screen lets staff add a member's number on any visit, confirmed with their PIN.

The ticks come next. One is required: "I accept the Terms and have read the Privacy notice." A second appears only if your program has wording for offers: "Send me offers and news in Wallet." It starts unticked, joining never depends on it, and offers go only to members who tick it. Leave that choice to the customer.

Then a 6-digit code arrives by email, the customer types it in, and the page says "You’re in." and offers the Wallet Card for their phone. No app is installed and no password chosen.

One setting can put a wait here on purpose. Under Who lets them in?, They join straight away issues the card "as soon as they finish the form." You approve each one means "They wait until your team says yes or no." The form then warns: "Use this only for a closed group, such as staff, students or invited partners. Everyone else is kept waiting for no reason."

The questions, the offers wording and who lets people in all belong to the program you send for review. Settle them for the counter before you submit: email, and the name only if the card shows it.

No published measurement was found of what a spoken invitation adds to a sign, or of one script against another. What follows is practice.

Keep it to two sentences, on the working assumption that a longer script gets dropped when there is a queue. An example for a café with a stamp card:

"We have a stamp card that goes in your phone's Wallet, no app needed. Scan this with your camera while I make your drink, and I'll put today's stamp on it."

Change the nouns for your trade and keep what it carries: the card lives in the Wallet, the camera opens it, and today's purchase counts. Say it while the customer is paying or waiting, so the wait for the emailed code falls inside time they were going to spend anyway. Say it to everyone, including the regulars you assume have joined.

Promise nothing the card will not show. In native.cards a new member's card starts empty: no Stamps or Points are given for joining, and staff cannot add stamps by hand. The card also needs a smartphone with Apple Wallet or Google Wallet. A customer without one is served as before.

Because the card starts empty, the first stamp or Points must come from a purchase, and the customer has just made one. Once the card is in their Wallet, staff scan it in the Scanner and press Record purchase, confirming with their own Staff PIN. The member leaves with something on the card. How to prepare your staff to run the program covers that screen.

There is evidence that this matters, with a twist. The American café study described in How soon should customers earn their first reward? ran a head-start test, separate from the one that timed finished cards: 108 customers were handed either a plain ten-stamp card or a twelve-stamp card with two stamps already on it. Both needed ten purchases; the head-start group finished in 12.7 days against 15.6. The same researchers also asked 65 travellers at a train station about a pizza chain's card. Two free stamps for joining made no difference to how likely they said they were to join: 5.9 against 5.8 on a scale of 0 to 10. That is 65 people answering about a card that did not exist, so hold it loosely. Its direction is still useful: a first stamp helps a member finish, and it does not win you the member.

Watch for the card itself. A customer can join and never add it. Such a member shows in Customers as No card: "No Wallet Card issued, so notifications cannot reach them". If the card will not add at that moment, the customer has still joined. An owner or admin can open them in Customers, read the Member ID on their record and type it into the Scanner, so the purchase still counts.

Signs are made under Programs, in QR Materials. A new one starts with Where will it go? and six places, the first of them Counter and tables ("Where people pay, wait or sit"). You pick a place, then a size. Pick a design comes next, with each design "drawn at this size with your own words and colours."

Those are five of 31 sizes, and Another size takes one you type in. For a counter sign the editor's own advice comes in two lines: "Stand it beside the card reader, facing the customer." "Paying takes seconds, so ask your staff to point to it." That is the product's advice, not a measurement.

Your own shop can measure which sign works, if each place has its own code, and the codes are already made. Once your program is approved and live, QR Materials opens with Ready-made QR codes, one for each place a code is used: General, Counter, Standee, Window, Receipt and Online. They all open the same join page, "but scans and joins are counted for each one, so Performance shows which place brings people in." Each is live from the start; there is nothing to send for approval. For a place the list lacks, such as a second counter or the delivery bags, Your own adds a category: "Shown in Performance, never to customers. Its code is live at once." The panel's one request is the one that matters: "Please use the correct QR category for accurate analytics".

To print a designed sign with one of those codes, the editor's five pages are Design, Words, Brand, QR code and Download. On QR code, choose One I already have and pick the code for the place the sign will stand. Everything that carries a code "shares its link and its count of scans", so a code belongs to one place: reuse it for more copies there, never for a sign somewhere else. The other choice, Its own new QR code, makes a code that native.cards checks before it goes live ("native.cards checks what a QR code opens before it goes live"). Keep it for a code that needs something a ready-made one does not have, such as a message shown before the join page.

Each code then has its own line under Campaigns, with the members who joined through it and how many came back. Which sign, link or post brings you the best members? reads those lines.

A printed code is permanent. If a sign has to stop working, the Performance tab of QR Materials has Pause… and Retire…: "Pause is reversible; retire is permanent." While a code is paused, anyone who scans a printed copy "is asked to try again later." Where the code leads can be changed without reprinting.