During a recent conversation on The Executive Outlook, Chris Tabb shared a truth many leaders need to hear before they invest in another AI initiative.
Most companies are not failing AI because they lack ambition. They are failing because they are starting in the wrong place.
Chris is a data and AI leader with nearly thirty years of experience across business intelligence, data architecture, governance, warehousing, modernization, and AI leadership. His journey began at Cognos in the 1990s, working in BI and reporting at a time when organizations were just starting to understand how data could support better decisions.
What makes Chris’s perspective valuable is that he did not begin with a purely technical lens. He started with a business lens. From the beginning, he saw data as a way to help decision makers become more effective. That early grounding still shapes how he thinks about AI strategy for business today.
For Chris, technology is never the starting point. Business value is.
Chris used a phrase that captures one of the biggest problems in modern data and AI work. He called it a technology beauty pageant.
It happens when teams build something impressive from a technical point of view, but the business value is unclear. A team may create the fastest pipeline, deploy the newest platform, or showcase the most exciting AI demo. But if it does not solve a real business problem, it is only impressive on the surface.
That is where many AI strategies lose direction.
Leaders want innovation. Boards want evidence that the company is moving fast. Teams want to work with exciting tools. So the organization starts with the most visible use case instead of the most valuable one.
Chris sees this mistake often. The chosen use case is usually too large, too complex and too slow to deliver. Stakeholders lose patience. Budgets get stretched. The team keeps working, but the business stops believing.
That is why Chris believes a strong AI adoption strategy should not begin with the shiniest idea. It should begin with the smallest valuable use case that can prove impact and build confidence.
Chris makes an important distinction between proof of concept and proof of value.
A proof of concept shows that something can be built. A proof of value shows that something is worth building.
That difference matters for every leader building an AI strategy for business. Many companies have created AI demos that looked promising but never became part of daily operations. They proved the technology could work, but they never proved that the business needed it enough to scale it.
He recommends starting with use cases that are quick to deliver, easy to measure and useful enough to create confidence across the organization. These early wins should not be random. They should become building blocks that unlock larger capabilities later.
In his view, good AI leadership means knowing the bigger vision but not trying to jump straight to it. The moonshot matters, but leaders must build the foundation before they chase it.
At the center of Chris’s thinking is the Context Layer.
He is very clear that the Context Layer is not one product, one vendor, or one technology. It is a logical layer that sits between the organization’s data and its AI systems. Its role is to give AI the meaning, terminology, relationships and business understanding it needs to produce reliable answers.
This is especially important because AI can sound confident even when it is wrong.
He explains that industries like legal and healthcare depend on specialized language. A word can carry a very specific meaning in one domain and a different meaning in everyday language. If an AI system does not understand that context, it may generate an answer that sounds correct but is not accurate.
That is why metadata, business glossaries, data catalogs, process definitions, semantic layers, ontologies and knowledge graphs matter more in the AI era than many leaders realize.
They are not just documentation. They are the meaning layer that helps AI understand what the business is actually asking.
Chris connects the Context Layer back to the earlier world of BI and reporting.
In the days of BI platforms like Cognos and BusinessObjects, semantic layers played an important role. They helped translate raw data into business-friendly definitions. They created a connection between the technical structure of data and the way business users understood it.
But during the big data era, Chris believes many organizations lost some of that discipline. Data was placed into lakes with the assumption that value would appear later. Companies collected more data, but they did not always model it for a clear purpose or connect it back to business's meaning.
That mistake is now becoming more dangerous with AI.
AI does not only need access to data. It needs access to the right meaning behind the data. Without that, the organization risks building powerful systems on weak understanding.
A modern AI strategy framework must therefore include context, not as an afterthought, but as a core foundation.
Chris also challenges the idea that governance slows innovation.
In his experience, governance becomes a problem only when it is brought in too late. When teams build first and ask governance or security teams to review everything at the end, delays are almost guaranteed. Issues around sensitive data, access, classification, anonymization, regulatory risk, data sovereignty and security controls suddenly appear when the project is already close to launch.
That creates frustration for everyone.
Chris’s advice is simple. Bring governance and security teams from the beginning.
When governance is part of the design process, it helps teams move faster, not slower. The right controls are built into the solution early. By the time the project reaches production, review becomes a smoother process because the important questions have already been addressed.
For any leader building an AI governance strategy, this is a critical lesson. Governance should not be the final gate. It should be part of the road that gets the solution safely into production.
Chris does not believe organizations can afford to keep waiting for the perfect time to modernize.
His point is practical. Competitors are already moving. Newer companies often have an advantage because they do not carry the same legacy infrastructure, technical debt and internal friction that older organizations must manage. They can move faster because they are not trying to untangle decades of systems and processes.
Chris used a simple line to explain the urgency. The best time to plant a tree was years ago. The next best time is today.
For business leaders, that means modernization cannot remain a future conversation forever. If organizations wait for the next tool, the next platform, or the perfect moment, they may never start.
But modernization should not mean changing everything at once. Chris recommends identifying business processes that can be improved quickly, choosing use cases that can show measurable value and building reusable assets that help future projects move faster.
That is how modernization becomes practical instead of overwhelming.
Chris returns again and again to a point many technology leaders recognize.
The hardest challenge is rarely the tool. It is the people and the process around the tool.
Business teams and technology teams often do not speak the same language. The business wants outcomes. Technology teams think in systems, pipelines, architecture and delivery. Both sides may be working hard, but they are not always working together.
He believes this gap must be closed if AI is going to create real value.
The best way to do that is to start with a small use case where business and technology teams can work closely together, measure progress and see the value of collaboration. Once people experience that alignment, it becomes easier to build trust for larger initiatives.
This is where AI strategies for business becomes less about models and more about operating rhythm. The organization must learn how to work differently if it wants AI to work meaningfully.
Across the full conversation, Chris Tabb is not offering another generic AI strategy framework. He is offering a grounded view shaped by decades of seeing what works and what fails in real enterprise environments.
His message to leaders is clear.
Do not start with the tool. Start with the business needs.
Do not chase the shiniest use case. Start with proof of value.
Do not treat governance as a late stage approval process. Build it into the design from the beginning.
Do not expect AI to understand your business without context. Build the metadata, semantic layer, catalog, glossary and governance foundation that make trustworthy answers possible.
Most importantly, do not confuse an impressive AI strategy deck with a working AI strategy for business.
The organizations that will lead with AI are not the ones with the loudest announcements. They are the ones building the foundations that allow AI to move from demo to decision, from experiment to production and from hype to measurable business value.
Want to hear more practical conversations with leaders shaping AI strategy, responsible innovation and enterprise transformation? Explore more on The Executive Outlook.
Listen in for business strategies and interviews with business experts to help you build a business and a life that you love.
About Us
