Every company has access to the same frontier models now. The same vendors are pitching the same platforms into the same use cases. If your AI strategy is “get the best model,” you don’t have a strategy. You have a subscription.
Context lives in your people, not your CRM. It’s in the deal review where a sales director notices a delivery lead quietly rephasing scope, and knows from years of pattern recognition that this deal is about to slip. It’s in the reason your best analyst can smell a bad forecast before the numbers say so. None of that shows up in a system of record. All of it shows up in how the work is done.
Harvard Business Review recently made this case with data most leaders haven’t seen. Researchers studied more than two hundred work patterns across more than fifty large enterprises, many of them Fortune 500 companies. Same industry. Same functions. Same systems. Execution still diverged, sharply. The variable wasn’t access to tools. It was context: the judgment calls, the trade-offs, and the unwritten rules about who gets looped in and when.
Which is exactly why the organizations pulling ahead right now aren’t the ones with the fanciest model. They’re the ones turning their own operating knowledge into something AI can use, often called a context layer.
Here’s the simple version. A context layer is a living record of how your business actually works: your definitions, your relationships, and your decision patterns, kept current and readable by AI in real time. Most companies already have something that looks similar: a data catalog, a semantic layer built for BI dashboards, a wiki of documented processes. The difference is what those were built for. A dashboard semantic layer only has to answer the handful of questions your reports already ask. A wiki gets written once and goes stale. Neither was built to answer a question nobody anticipated, which is exactly what an AI agent willask it. A context layer is built to stay current and to hold up under unbounded, unpredictable use.
What this looks like in practice
Last month at Stanford’s Institute for Human-Centered AI, Airbnb’s Global Head of Employee Experience, Iain Roberts, described how his team is doing exactly this. Two moves stood out.
First, they killed the PDF. “Slides and PDFs are the enemy of intelligence,” Roberts said. PDF files are essentially a set of instructions for rendering pixels on a page. Critical callouts might be reduced to “bigger text at position (x,y) rather than “heading”. The lack of the consistent hierarchy is pumping up token cost and pushing out results not anchored in a cohesive connective thread. His leadership team now writes in markdown: plain text that’s machine-readable, version-controlled, and easy to verify. It’s a small habit change with a large consequence. You cannot feed an AI system your organization’s thinking if that thinking is trapped in a slide deck.
Second, they’re measuring how well strategy travels across the organization. Airbnb ingests recordings from every leader’s town hall, from VP to C-suite, and tracks two things: how long it takes a message to move from “here’s where we’re going” to “here’s what that means for your work,” and where a function’s actual behavior diverges from the strategy it just heard. That’s organizational context, captured continuously, instead of assumed.
None of these moves required a new model. They required deciding that the organization’s own operating knowledge was worth capturing in a form AI can use to dramatically increase the accuracy and value of the model’s output.
The work ahead
Here’s the uncomfortable part for most leaders: building this isn’t a data project you can hand entirely to IT, and it isn’t a prompt-engineering exercise you can hand entirely to your AI team. It’s an architecturedecision. To fully realize the potential, it requires taking a step beyond Airbnb’s initial steps described above. It touches how you document, how you define your business terms, who owns which definitions, and how that knowledge gets delivered to every model and agent your teams pick up next.
Get that foundation wrong and every AI pilot you run will look impressive in a demo and plateau the moment it hits real work. That’s exactly the pattern researchers keep documenting: promising pilots, then nothing that lasts.
The moat was never the model. It’s the map of how your business works, and whether you’ve bothered to draw it in a language your AI can read.
If this is on your roadmap, here’s where Unify’s Data and Analytics practice usually starts: a short assessment of where your organizational context currently lives, across systems, documents, and the heads of your senior people, followed by a scoped pilot that maps one high-value workflow into a usable context layer for a single AI use case. Reach out and we’ll walk through what that looks like for your organization.
Shift happens. And sometimes, it happens because of the people we assumed were already on board.
About The Shift Series
Shift Happens is a series exploring how organizations can turn disruption into direction. We write about the real, human side of work, where change, technology, behavior, and leadership collide in ways no framework fully captures.
Every article follows one of the five currents that shape modern work:
The Human Side of Transformation, the heartbeat beneath the strategy.
Change Management as the Missing Discipline, the discipline hiding in plain sight, quietly determining who succeeds.
Technology, Tools + Human Behavior, the space where logic meets instinct, and where most rollouts live or die.
Organizational Structure, Power & Governance, the lines, ladders, and tensions that decide how work truly flows.
Leadership Micro-Shifts, Governance & Operating Models, the small shifts that create disproportionate impact.
We combine lived experience with practical insight. The kind you can apply the same day, not someday.
Shift happens! But with the right mindset, it happens through you.
If your organization is navigating a shift in technology, structure, or culture and needs practical, human-centered support, reach out.
This is the work we love! And the work we do best.
