REAL Talk With Sam Holcman

Welcome to Real Talk with Sam Holcman - where we stop confusing activity with progress and start talking about how businesses actually work.

Today I want to talk about something touchy in the Business Architecture world: the difference between belonging to a club and practicing a discipline. Specifically, the world of the Business Architecture Guild® and its BIZBOK® Guide on one side… and BACOE, the Business Architecture Center Of Excellence, on the other.

Many CEOs excel in vision, leadership, and financial acumen - yet remain only vaguely familiar with the disciplines of Enterprise Architecture (EA) and Business Architecture (BA). They hear the words. They see the frameworks. But they do not always connect with the business value that drives strategy, change, and decision-making. That gap can be dangerous.

Enterprise Architecture (EA) is shifting decisively from theory-heavy, exam-driven certification to demonstrated, practice-based competence that delivers measurable business outcomes.

Organizations have been stockpiling data for years, expecting that one day it will become a strategic asset. With generative AI, that moment has arrived - but without disciplined data practices, the promise quickly turns into frustration. The differentiator is no longer access to powerful models; it is the ability to shape, govern, and trust the data that feeds them, which is exactly the gap the AI Data Modeling Masterclass is designed to close.

Let us start with the uncomfortable truth: large language models - ChatGPT, Gemini, Claude, Perplexity, and their peers - are built on a foundation that looks suspiciously like “copyright infringement as a service.” These systems could not exist without consuming massive quantities of human-created text, images, and code. In another industry, that would be called derivative work. In AI, it is called innovation.

Business Architecture (BA) is undergoing the same transformation: away from framework-centric, multiple-choice exams and toward practitioner-based competence proven through real work, with BIZBOK/CBA®-style certification as a well-known example of the older model, not the only one.

There is something absurd happening in enterprise architecture and business architecture today, and it is time we called it out.

People are passing multiple-choice exams… or cramming through a sixteen-hundred-page framework manual, and then being told, Now, go modify it. Customize it for your organization.

Modify it? Before you have ever used it? Based on what? That is not mastery. That is madness.

Most EA principles sound good but do not change a single decision because they are not testable.

To make them real, convert each principle into a SMART goal plus 1–3 hard metrics. 

Now you can wire that into governance (architecture reviews, portfolio decisions, roadmaps) and report progress quarterly. At that point, it stops being a poster on the wall and becomes a scoreboard the business can hold EA accountable to.  Let us explain why Architecture Principles faded away in the 1990's - let us not go backwards in time.

This Broadcast is a review and analysis of the Jan 22, 2026 - by Koenraad Schelfaut and Andrew Long CIO Article titled How to get your enterprise architecture ready for agentic AI.  This article misses the central - and fatal - flaw in its reasoning: it assumes that enterprise data is reliable enough to empower agentic AI when, for most organizations, it is not.  A concise 90-day, EACOE EA-anchored roadmap you can use with CIOs and sponsors is provided.

“If you are worried about the cost of going for it, just calculate the cost of staying where you are."

We have all been there - staring at an opportunity that feels just out of reach. Maybe it is the next career move, a new certification, launching that business idea, or stepping up into leadership. The first thought that comes to mind? “Can I afford this?

But here is the truth most people never calculate: staying where you are, has a cost too - it is just hidden.