ON DEMAND • Podcast Episode

Why 95% Of Enterprise AI Never Ships: Scaling AI Transformation Beyond Pilots

Emanuele de Santis, management consultant at AlixPartners, joins Better Tech to discuss AI transformation, data governance, change management, digital ecosystems, and why successful enterprise AI transformation needs strong business pain points before new tools.

FEATURED GUEST

Emanuele de Santis

Management Consultant, AlixPartners

Our featured guest

Emanuele de Santis brings a transformation-focused perspective shaped by a decade of helping clients through large-scale cost and digital transformation programs across industries. He explains why organizations often struggle to improve margins when processes remain fragmented, systems are siloed, and leadership lacks visibility into where inefficiencies actually sit.

He also covers:

Why AI transformation starts with disciplined data
How change management determines whether new systems are actually adopted
Why agentic AI still struggles when human intervention becomes excessive
How semantic layers and digital twins make enterprise data easier to use
Why business value and aligned incentives matter before scaling AI

Why this conversation matters now

Organizations are investing heavily in AI, but many are discovering that fragmented data, weak governance, misaligned teams, and poor adoption can limit the value of even the best technology.

This episode explores how organizations can approach enterprise AI adoption more effectively and what needs to happen before scaling AI in the enterprise becomes realistic.

01.

How do we move from scattered AI pilots to scalable AI transformation?

02.

Why does clean, governed data matter before AI can deliver value?

03.

How do we align IT and business transformation teams around the same goals?

04.

What happens when a human-in-the-loop becomes human-in-the-middle?

05.

How can semantic layers make enterprise data easier to query and act on?

06.

What should leaders prioritize before investing in large AI programs?

What you'll learn in this episode

After unlocking the full podcast, you’ll get expert insights on AI transformation, AI data governance, change management, semantic layers, and scalable enterprise AI adoption.

01.

Why large transformation programs often fail when teams lack shared incentives

02.

How AI change management goes beyond communication and into adoption, sponsorship, and behavior

03.

Why AI initiatives need clean, harmonized, well-governed data first

04.

How semantic layers and digital twins make business data easier to use

05.

Why some agentic AI initiatives create more human work instead of less

06.

How AI-assisted software development can speed up deterministic solutions

07.

Why strong business pain points should come before AI technology choices

Who this episode is for

CEOs and business leaders CIOs and technology leaders Digital transformation leaders Data and AI leaders Finance and operations leaders Growing businesses planning AI transformation

Our host

Umair Javed

CEO & Founder

Umair Javed is an AI transformation leader with extensive experience helping organizations adopt and scale AI, including Generative AI and Agentic AI. As CEO of tkxel, he has worked with Fortune 500s, SMBs, and startups worldwide, guiding them through AI-driven transformation, digital modernization, and long-term business growth.

A preview of the key takeaways

01.

AI transformation starts with disciplined data

Emanuele explains that organizations cannot expect reliable AI outcomes without clean, harmonized, well-governed data. A strong semantic layer can reduce ambiguity and make it easier for business users to access the right context, making AI data governance a foundational part of successful transformation.

02.

Change management is about adoption, not messaging

Transformation can fail even when the technology works. People may resist because they fear job impact, dislike new workflows, or simply prefer familiar systems. Effective AI change management requires clear value, aligned incentives, sponsors, and internal champions to make adoption stick.

03.

Agentic AI still needs careful boundaries

Many organizations have learned that early agentic AI initiatives can require so much human intervention that the economics stop making sense. Emanuele argues that companies are now becoming more selective about where autonomy actually adds value as part of a broader AI transformation strategy.

04.

Semantic layers can turn fragmented data into usable business context

Modern data platforms can connect information from separate systems and create business-friendly object models that make analytics easier to query, combine, and apply inside workflows. These foundations can make enterprise AI transformation easier to scale across functions.

05.

The right starting point is a painful business problem

Emanuele’s advice is simple: start with a clear business pain point, attach measurable value to solving it, and align IT and transformation teams around the same incentives before scaling. This creates a stronger foundation for enterprise AI adoption and long-term transformation.

Why tkxel is hosting this conversation

 tkxel helps growing businesses move from AI experimentation to practical AI transformation through data readiness, AI strategy, software modernization, and workflow redesign.

This episode reflects a challenge many teams face today: AI can accelerate analysis and automation, but fragmented data, weak adoption, and misaligned incentives can stall progress. Emanuele’s perspective adds context for leaders trying to build the foundations, governance, and organizational alignment needed to turn AI investment into measurable business value and support scalable enterprise AI transformation.

WHERE TKXEL BRINGS CONTEXT

1 AI transformation strategy

2 Data readiness

3 Workflow modernization

4 Software engineering

RECOGNIZED BY

Key concepts covered in this podcast

AI transformation

How organizations move from isolated pilots to business-wide value.

AI data governance

Why clean, consistent, controlled data is foundational to AI.

Semantic layers

How business-friendly data models make enterprise information easier to use.

Digital twins

How connected data objects can represent real business entities and processes.

AI change management

Why adoption, incentives, and internal champions matter as much as technology.

Agentic AI

Where autonomous workflows can help, and where human intervention still creates friction.

Business value

Why AI initiatives should begin with measurable pain points and outcomes.

Frequently Asked Questions (FAQs)

Is this episode only for large companies? Expand FAQ

No. While many examples come from large transformation programs, the lessons are also relevant for mid-market companies dealing with fragmented data, siloed systems, and AI adoption challenges.

What is the core focus of the episode? Expand FAQ

The episode focuses on AI transformation, AI data governance, semantic layers, agentic AI, change management, and how organizations can scale digital initiatives around measurable business value.

Can this help teams struggling to scale AI pilots? Expand FAQ

Yes. The conversation is useful for teams dealing with poor data foundations, adoption barriers, unclear ownership, or AI initiatives that require too much manual intervention. It also provides practical context for organizations trying to move from pilots toward scaling AI in the enterprise.

Why is this podcast gated? Expand FAQ

It is offered as an on-demand resource for leaders researching AI transformation strategy, data strategy, digital modernization, and scalable AI adoption.

Access the full podcast

We'll only use your details to send this resource and relevant insights from tkxel.