PH Leads APAC in AI Ambition — But Fragmented Data Is Quietly Killing ROI

Your team is using AI tools. You are paying the subscriptions. But the results still feel underwhelming. A new study from Boomi and research firm Omdia might explain why — and the culprit is not the AI itself.

PH Is Serious About AI — But the Foundation Is Shaky

73% of Philippine firms have launched active AI initiatives, according to the Boomi-commissioned Omdia study covering the Asia-Pacific region. That is a strong number. Even more striking: 50% have already set KPIs to measure AI success — the highest rate in all of Asia-Pacific.

But here is the problem buried inside the same study: 97% of those firms agree that AI requires sharper focus on data quality and governance. That is basically everyone admitting the foundation is not yet ready.

What 'Fragmented Data' Actually Means for Your Business

Fragmented data is not a technical concept — it is what happens when your customer records live in one app, your sales history in another, your inventory in a spreadsheet, and your chat conversations in a third platform. None of them talk to each other cleanly.

When AI runs on top of this setup, it pulls from incomplete or contradictory sources. Outputs become unreliable. The dashboard looks impressive, but the actual business results disappoint.

Boomi's Chief Technology Officer David Irecki put it plainly: "If data is being moved or transformed through unmanaged channels, the organization does not know where the data came from, whether it's current, whether it has changed, or who has access to it."

The study calls this the shadow integration problem — when teams move data through makeshift channels (manual CSV exports, copy-paste, WhatsApp file forwards) outside any governed system. It is extremely common among small and mid-sized businesses. And it quietly poisons AI results.

The Gap: Big Plans, Thin Infrastructure

The numbers paint a clear picture of where PH firms stand:

• 90% believe AI-enabled automation will reshape their business processes within two to three years
• 93% plan to shift toward unified, enterprise AI-ready data platforms
• Only 50% currently use any integration platform approach — meaning half are still running disconnected systems

The ambition is real. The infrastructure gap is also real.

What You Can Fix This Week

The full solution — a unified data platform — is a multi-month project. But there are practical steps to reduce the damage now:

• Audit your data sources. List every tool your business uses that holds customer or operations data. Count how many are disconnected from each other. That list is your fragmentation map.
• Pick one integration layer. Tools like Make, Zapier, or n8n (all with free tiers) can connect common business apps without writing code. Even one working connection reduces fragmentation meaningfully.
• Define AI inputs clearly. Before asking any AI tool to summarize, recommend, or report — specify exactly which data source it should draw from. Garbage in, garbage out applies to every AI tool, regardless of cost.
• Start with one clean dataset. Pick your most valuable data — customer records or sales history — and clean that first. AI built on one clean, reliable dataset consistently outperforms AI built on five messy ones.

The Honest Takeaway

PH businesses are not behind on AI enthusiasm — they are ahead of most of the region. What is lagging is the unglamorous infrastructure work that makes AI outputs trustworthy. Companies that address this gap early will get measurably better returns from the tools they are already paying for.

Source: BusinessWorld: Fragmented data holding back PHL firms' AI success, Boomi says