Across APAC, most executives I meet have no shortage of ambition. They talk about hyper‑personalisation, omnichannel journeys and AI‑powered service. Yet when you sit with the agents, case workers or relationship managers who are meant to bring that vision to life, a very different reality emerges.
The real CX execution gap sits between strategic intent and operational accessibility. Leaders may have a clear North Star, but the people closest to customers often lack one simple thing: usable data at the moment of need. In fact, 82% of C‑suite leaders rank digital transformation as a high priority, yet only 3% say they’ve achieved a fully integrated, agile digital ecosystem.
In APAC, this is playing out as what I call “pilot purgatory”. Organisations launch sophisticated AI proofs‑of‑concept, but because the underlying data remains fragmented, they struggle to scale beyond isolated use cases. The result is an impressive slide deck, but very little that changes the customer’s experience on a Tuesday afternoon.
How silos silently destroy revenue and relationships
The root cause is technology overload. The average enterprise now runs nearly 1000 applications, with only around a quarter properly integrated. When systems don’t talk, data quality deteriorates – a problem that, according to IBM, leads more than a quarter of organisations to lose over US$5 million annually, with 7% losing US$25 million or more.
Consider the renewal journey. A customer approaches a contract end-date, but their success data (usage and outcomes) sits in one system, while billing and contractual commitments sit in another. Without a unified view, the account manager cannot see which promises were kept or where friction emerged. The conversation defaults to a discount rather than an insight-led upsell. Multiply this across thousands of accounts, and the impact on growth is profound.
The same friction exists in support journeys. A customer expects to be recognised as the same person whether they start on a portal, move to chat, or speak to a human. If every channel creates a new case because those systems cannot share context, the customer is forced to repeat themselves and likely causing frustration. Over time, this experience erodes loyalty far more than a single technical issue ever could.
AI is not a strategy – and can amplify the problem
Into this fragile environment, we are injecting generative AI. A dangerous misconception is that “AI is the strategy.” It is not. AI is a capability that will either elevate your CX or, if deployed on top of fragmented data, scale your broken experiences.
We see organisations rushing to stand up AI assistants that are, in effect, empty shells: impressive interfaces connected to incomplete data. The outcome is predictable: contradictory information and interactions that damage brand trust. Conversely, when AI is grounded in a robust data foundation, the upside is significant. In fact, 58% of executives report that responsible AI practices, built on governed, high‑quality data, improve their ROI and efficiency.
The non‑negotiable foundations for AI‑ready CX
To move beyond the execution gap, three non-negotiable foundations must be in place:
-
A unified agreement and knowledge foundation: A governed source of truth that centralises critical data, including the agreements that define your customer obligations. This powers consistency across chatbots, workflows, and humans.
-
Contextual intelligence: Systems must understand a customer’s intent and history in real-time. This context must follow the user seamlessly across channels and business units.
-
Closed-loop governance: Clear policies defining when automation is appropriate and when human escalation is mandatory. Without trust in the data, AI experiments will never handle meaningful volume.
A pragmatic roadmap out of pilot purgatory
Breaking these bottlenecks does not require a five-year overhaul, but it does require sequencing.
-
Step 1: Map the end-to-end customer journey. Identify a specific item (like disputes or renewals) and map every system and hand-off involved. This reveals exactly where you are asking customers and employees to work too hard.
-
Step 2: Establish a source of truth. Rather than trying to fix every silo at once, cleanse and curate the foundational data sets that matter most to that specific journey.
-
Step 3: Prioritise agent-assisted experiences. Use AI to remove duplicate entry and surface agreement context in real-time. This approach of keeping a human in the loop reduces risk and builds internal confidence faster than jumping to full automation.
Only then should you scale into production-grade automation. By this stage, you will have a direct line of sight between AI investment and measurable business outcomes.
Over the next three years, the winners in APAC will not be those who adopted AI the fastest. They will be the leaders who took the time to fix their data, break their silos, and close the execution gap between the CX strategy on the wall and the experience a customer actually feels.