What's NEXT in Marketing: AI-driven loyalty and Southeast Asia's $2T opportunity
Personalisation has moved from buzzword to profit engine. Boston Consulting Group forecasts that, over the next five years, $2 trillion in revenue will shift to companies that deliver truly personalised experiences and communications. That's not a trend. That's a reallocation of market share to brands that get closer to the individual customer.
Recent panels in Bangkok and Taipei with retail leaders and representatives from Google Cloud, Airship, Fifty-five, and NTT DATA all landed on the same point: personalisation directly impacts profitability. The businesses that act now will see it first. The ones that wait will pay for it later.
Omnichannel is the business; personalisation is the experience
Omnichannel is your internal goal: a connected operation across stores, app, web, and media. Personalisation is what the customer feels: "This brand gets me." They're two sides of the same coin. If you want higher conversion and margin, make every touchpoint feel 1:1.
Good retailers don't chase trends. They shape behaviour. They use personalisation to increase spend, frequency, and lifetime value-without overspending on generic discounts.
Modern loyalty needs a connected system
Loyalty works when it rewards the exact behaviour you want. The upside sits in your current customer base, hidden in plain sight. But it only shows up when four elements move together.
- Data: Connected, clean, privacy-safe, and real-time across channels.
- Insight: Always-on models that predict intent, value, and next best action.
- Loyalty: Immediate recognition and rewards that feel meaningful.
- Action: Instant execution across POS, app, email, ads, and service.
Turn the DIAL together or expect leakage. Fragmented systems equal wasted spend.
AI is accelerating 1:1 at scale
Retailers are moving past segments to true individualisation: what to offer, when, and why. At Eagle Eye, Eagle AI processes billions of interactions per minute to make these micro-decisions in real time. The more interactions it sees, the sharper the targeting and the lower the cost to win each action.
Personalisation is also widening beyond promotions: personalised cooking plans, health and wellbeing nudges, even charitable preferences. Done right, relevance feels helpful-not pushy.
Data foundations without the multi-year slog
The old approach was "build the perfect data platform, then deliver value." That's slow and expensive. A smarter path: pick one problem with real financial impact, run a rapid proof of concept, prove lift, then expand. Small steps, fast pace.
This approach matches how AI evolves. Three-year roadmaps lag. 90-day cycles win.
What the results look like in practice
Ten years of personalisation work has compounded. "We started using AI to personalise promotions in 2015," said Eagle Eye CEO Tim Mason. "We have seven algorithms that work in concert to create a promotion programme."
Adding transformer neural networks on top of those algorithms had an immediate effect: "The take-up rate of suggested offers has gone up by 20%," Mason said. "Using a large language model, it's basically able to look around corners."
Woolworths Group re-platformed its loyalty engine to run in real time. Former CEO Brad Banducci described it as "instantaneous," with full history reconciliation and no constraints on offer design. Tesco's history tells a similar story: Clubcard helped lift share from 23% to 33%, and recent "Clubcard Challenges" delivered strong responses to light personalisation and gamification, contributing to Tesco's highest market share since 2016.
Southeast Asia's edge
Mobile-first consumers, super-app behaviour, and digital payments create ideal conditions for real-time personalisation. With proven results globally, SEA retailers can compress learning curves and move straight to outcomes. The $2T shift won't wait for laggards.
Why relevance beats blanket discounts
Spraying the same offer across your base is expensive and dull. Relevance lowers your media and promo burn while lifting conversion and margin. You spend less to get more.
A 90-day plan to get moving
- Pick one metric to move: repeat purchase rate, average order value, or category penetration.
- Form a small squad: marketing, data science, engineering, store ops (or CX).
- Connect just enough data: identity resolution, SKU-level transactions, consent status.
- Ship a real use case: weekly 1:1 offers for a high-value category with control groups.
- Execute everywhere: POS, app, email, paid media suppression/boosting.
- Measure lift fast: redemption, incremental sales, margin impact, cannibalisation checks.
- Scale what works, kill what doesn't, and expand audiences and categories.
- Lock in guardrails: frequency caps, fairness rules, privacy and governance.
Execution details that matter
- Real time is non-negotiable. Think thousands of API calls per second at checkout and in-app.
- Offer science > bigger budgets. Calibrate incentive by predicted value and price sensitivity.
- Creative still counts. Plain language and clear value beat fancy templates.
- Close the loop. Train models with outcomes, not just clicks.
Bottom line
The technology exists. The business case is proven. Over the next five years, revenue will shift to brands that deliver individual experiences at scale. The only question left: how fast can you execute?
Boston Consulting Group on personalisation and growth
Google Cloud Vertex AI
If your marketing team needs practical upskilling on AI personalisation and measurement, explore the AI Certification for Marketing Specialists.
Your membership also unlocks:
