Rolling out a loyalty app across 4 markets sounds like a tech problem. It's actually a marketing problem. The app works fine. Getting people to download it, use it, and keep using it is where most rollouts fail. Here's what we learned launching the DON DON DONKI app across Singapore, Malaysia, Thailand, and Hong Kong.
Most brands assume that if the app exists and the offer is good, downloads will follow. They won't.
Downloads are driven by in-store staff behaviour, not marketing. If cashiers don't proactively offer the app at checkout, download rates collapse. No amount of Instagram ads or email blasts changes this.
"If cashiers don't proactively offer the app at checkout, download rates collapse."
We saw this clearly in the DONKI rollout. Stores with trained staff who said "Would you like to scan your loyalty points?" at every checkout converted at 3.2x the rate of stores without the protocol. This wasn't a product problem. It was a training and process problem.
Most brands send too many, too generic, and at the wrong time. "Check out our latest deals!" at 2pm Tuesday gets ignored. "Your 50 points expire in 3 days" at 9am Sunday gets opened and acted on.
The difference isn't messaging tone. It's relevance and timing. Personalisation isn't optional. It's the difference between 8% open rates and 35%.
During the DONKI launch, we tested three strategies across the four markets:
Behavioural triggers won by a wide margin. Segmented campaigns with country-specific timing came second. Generic blasts performed worst.
We made three mistakes that showed up immediately in user feedback and 1-star app store reviews.
If it takes more than 3 taps to use a reward, most users won't bother. We designed the flow with 5 taps. Checkout time is stressful already. Adding friction to the loyalty experience meant people just stopped using the app.
Onboarding asked for date of birth, phone number, and ID number. Drop-off at step 2 was significant. We were collecting data we didn't need to collect at signup.
App store ratings are a UX signal. If your 1-star reviews all say "too complicated" or "can't find my points", that's a UX audit task, not a PR crisis. We watched app ratings climb from 3.1 to 4.3 stars across all four markets after these two changes.
One launch strategy does not work across 4 markets. The app is the same. The marketing approach must be different.
Highest digital literacy. Fastest to adopt. Most responsive to push notifications. QR code adoption was immediate. Incentive-driven mechanics worked well. The app felt like a natural addition to the shopping experience.
WhatsApp-first market. App downloads were slower, but WhatsApp broadcast adoption was high. We pivoted focus to driving users to WhatsApp broadcasts instead of push notifications. Loyalty engagement happened there instead.
Influencer seeding drove the highest initial download spike. Organic word-of-mouth through local influencers outperformed all paid channels. In-store education was also critical because app navigation conventions weren't assumed.
App adoption was slower. QR code scanning at point of sale was the most effective entry point. Direct app download incentives performed worse. Trust-building and staff recommendations mattered more than promotional mechanics.
You probably don't have a regional app launch in your roadmap. But these principles apply to any loyalty mechanic or digital touchpoint.
The biggest insight from the DONKI rollout wasn't about technology. It was that adoption is a behaviour problem, not a product problem. Fix the behaviour first. The product just has to stay out of the way.
A Page Audit shows you exactly where signups drop off, which messaging underperforms, and what your competitors are doing differently. Delivered in 48 hours. From SGD $99.
Get an audit →