When the Floor Is Lava, Learn to Dance — How North Highland Moved Into the AI Era
A first-person account of how a global consulting firm built, deployed, and scaled AI across 1,400 people — not by buying a tool, but by becoming a firm that runs on one.
Chapter I: The Vault
How a decade of institutional memory became our most important AI asset.
Our AI story doesn't start in the 2020s. It starts about a decade earlier, in the day-to-day work of consulting. Over years and thousands of projects, we built something you can't buy: a living account of what we actually knew. Playbooks that evolved from contact with real clients. Templates new teams could pick up the same day. A running record of what worked in a merger, a service redesign, a technology migration, an op model transformation.
The reason AI worked for us is that we had something worth feeding it.
Principle we operate by
AI extends people. We gave people a way to work faster, so they could spend more time on judgment, creativity, and the client work only a human can do.
Client Application: Data & Knowledge Foundations
When clients ask where to start on AI, the honest answer is: not with a tool, but with the work of surfacing, structuring, and protecting the knowledge you already own. AI amplifies what's there. Nothing more.
Now ask yourself
- If you fed your best knowledge to AI tomorrow, what would come back?
- Do you know where your best thinking actually lives?
- Who decides what "good" looks like for your knowledge?
Chapter II: The Center
Why you can't scale AI without somewhere — and someone — to hold the standard.
You can't have an AI strategy without someone whose job it is to think about AI. Not a committee. Not a subgroup of the technology council. A dedicated center with real authority, real budget, and real accountability for outcomes.
Principle we operate by
Centralize the expertise. Distribute the capability. The center holds the standard — but the standard has to travel.
Client Application: AI Centers of Excellence
Most organizations that struggle with AI adoption have no clear home for it. Someone owns the tools budget. Someone else owns the training. A third team owns the use cases. Nobody owns the outcome. Establishing a center — even a small one — with explicit authority over AI quality and direction changes the dynamic.
Now ask yourself
- Who owns AI outcomes in your organization today — not the tools, the outcomes?
- Is that ownership clear enough for someone to act on it?
- What would it take to give someone real authority over AI quality?
Chapters III–VIII: Available in the Full Book
The Platform, The Decisions, The Economics, The People, The (non) Conclusion, and The Whole Story are available by download.