The Missing Element: Making the Invisible Visible

Welcome back to Connecting.The.Dots — a little ahead of schedule. But hey, I had this article on my chest, and really wanted to share it as soon as possible.

In the 1st edition, I introduced the "Missing Element" that separates GBS 3.0 leaders from survivors. So now it's time to unveil what the missing element is. That missing element isn't a checklist or a tool — it's simply the ability to observe and deeply understand why things fail. Sorry to disappoint you: no complex algorithms or shiny tools. But it's never been more important to make the invisible visible than in an era where AI is creating the pathway for GBS 3.0 transformation.

This edition is a deeper reflection into that missing element — how do we make the invisible visible, rendering blind spots clear so you can finally understand what's breaking your AI investments.

Today, we're making a new (but at the same time, well-known) component visible — Data Silos — which I'll reframe as "Data Hedges." This reframing matters because once you see silos as hedges — natural, necessary boundaries that grow into barriers when left unattended — you can finally diagnose why your AI hallucinates, why your digital twins remain incomplete, and why the Intelligence Triad can't amplify.

Your AI use cases aren't failing because of the technology. They're failing because you can't see what's breaking them.

The million-dollar question: why does AI fail?

I've watched numerous organizations in the past few months invest heavily in Generative AI, surfing the hype curve to the max — sophisticated models, talented teams, ambitious roadmaps. Yet six months later, they're struggling to move beyond pilots and proofs-of-concept.

The culprit isn't the AI capability. It's not the talent. It's not even the strategy.

It's the Data Hedges. Let's unpack this.

The reflection came to me from a recent leadership training, while we explored organizational silos. And I had an aha moment: this is exactly what we are facing on AI, Data and Technology adoption across our organizations.

Hedges vs. silos: understanding the difference

In the digital ecosystem, silos are not structurally intrinsic. They're not deliberately designed to cause problems. Instead, they emerge naturally — just like hedges in a garden.

Think about it: when you plant a hedge, you're creating a boundary. You're establishing order. You're defining controlled perimeters. This is necessary and good. Different departments need different systems. Finance operates differently from HR, which operates differently from Supply Chain. These boundaries are intentional and valuable.

But here's what happens: hedges grow. Naturally. Continuously. And if left unattended, they transform from useful boundaries into impenetrable walls — silos — where data is trapped, connections are lost, and visibility disappears.

• The line of sight that once existed? Gone.
• The data that should flow freely? Fragmented and stuck.
• The business intelligence you need? Locked behind overgrown hedges.

The digital-twin problem: when hedges block reality

Consider a standard procure-to-pay process running across three different systems: a procurement system handling requisitions and purchase orders, an ERP managing approvals, receiving, and inventory, and a finance system processing invoicing and payments.

To truly understand this process — to create its digital twin — you need to connect the data across all three systems. You need to see:

• How long requisitions wait in procurement
• Where approvals bottleneck in the ERP
• Why invoice matching fails in finance
• Which suppliers cause the most friction

But when data hedges are left untrimmed, this becomes impossible. Each system holds a fragment of the truth. Each database speaks a different language. Master data definitions vary. Reference data conflicts. And the complete picture — the digital twin of your actual process — remains invisible.

This is why your AI hallucinates. It's not getting a complete view. It's getting three fragmented perspectives and trying to make sense of a process it can't fully see.

Master data & reference data: the foundation that breaks

Here's where it gets critical: Master Data and Reference Data are the foundation of any AI-driven transformation.

• Master Data defines your business entities — customers, suppliers, products, employees, locations. When master data is inconsistent across systems, your AI doesn't know it's looking at the same customer in three different databases.
• Reference Data provides the context — currencies, units of measure, categories, hierarchies. When reference data varies between systems, your AI can't make accurate comparisons or draw meaningful insights.

Without clean, connected master and reference data, you're not feeding AI. You're starving it.

Think of it this way: if your organization were a body, master data is the bloodstream. Reference data is the nervous system. When hedges create silos, you're essentially cutting off circulation and nerve signals to different organs. The body still functions — barely — but it can't perform optimally. It certainly can't amplify intelligence.

The one-silo-at-a-time trap

Here's how most organizations approach this problem: they identify a silo. They fund a project. They "fix" it — usually by cleaning data in that one system or building a point-to-point integration. Then they move to the next silo. And the next. And the next.

Meanwhile, the hedges keep growing. New systems get added. New boundaries emerge. New silos form faster than old ones get resolved.

This is the organizational equivalent of bailing water from a sinking ship — one cup at a time.

The problem isn't lack of effort. It's lack of systemic thinking. You can't trim one hedge and expect the garden to thrive. You need a comprehensive approach to maintenance and connectivity.

The path forward: from invisible to visible

Most organizations can't diagnose their AI failures because they can't see the hedges. They're trapped in the one-silo-at-a-time approach, watching their hedges grow faster than they can trim them. They're investing millions in AI while their data remains fragmented, disconnected, and siloed — but they don't understand why.

The result? AI that hallucinates. Digital twins that remain incomplete. Intelligence that can't amplify because it can't flow. And leadership teams that can't explain what's going wrong.

But now you can see it.

You understand that silos aren't the problem — they're natural boundaries that require maintenance. You recognize the hedges analogy. You can diagnose why your AI fails: it's starving for connected data, trapped behind overgrown boundaries, unable to see complete processes.

This visibility is the first step. You can't fix what you can't see. But seeing isn't solving.

There's another way — one inspired by nature's own solution to connectivity. A way to create data ecosystems where information flows freely, where systems connect naturally, and where your AI can finally access the complete picture it needs to deliver on its promise.

More on this in Edition #3.

This week's action: map your hedges

Until then, I want you to do one thing: map your hedges.

1. Look at your critical processes. Identify where they cross system boundaries.
2. Document where data gets trapped, where definitions conflict, where master data breaks down.
3. Make the invisible visible in your own organization.

Don't try to fix anything yet. Just see the hedges for what they are. Because you can't trim what you can't see.

Follow the conversation   #ConnectingTheDots   #GBS3_0   #DataStrategy   #AITransformation   #DigitalTwin

If this resonated with you, share it with a colleague struggling with AI implementation. Sometimes the diagnosis is more valuable than a premature prescription. New here? Subscribe and catch up on Edition #1 — it's going to be an interesting journey ahead.

This content is based on the author's independent reflections and is not associated with any company or organisation.