The transformation worked. The strategy was clarified. Capabilities were mapped. Gaps were identified and prioritized. The leadership team had a shared language for the first time. Decisions that used to take months happened in weeks. Everyone agreed: this was how the organization should work.
And then someone asked: "Who maintains this?"
It's not a glamorous question. It doesn't appear in strategy presentations or board reports. But it's the question that determines whether everything the organization built persists or decays.
Because a shared vocabulary is not a document you create once and file. It's a living artifact. The strategy will evolve. New capabilities will be needed. Old ones will become irrelevant. The external environment will shift. If nobody is responsible for keeping the vocabulary current, it will drift from reality. Within a year, the business model describes something the organization no longer operates. Within two years, the capability map lists capabilities that have been reorganized beyond recognition. Within three years, the shared language that made the transformation possible is an artifact in a folder that nobody opens.
This is the story of most organizational transformation: not failure at the doing, but decay after the doing. Not a dramatic collapse, but a quiet drift back to the old confusion.
Building Is Intellectual. Sustaining Is Organizational.
Building a shared vocabulary is an intellectual exercise. You need architectural thinking — the ability to see structure, make design decisions, maintain consistency. That skill set is well understood.
Sustaining a shared vocabulary is an organizational exercise. You need engagement, governance, political navigation, and persistent attention. That skill set is different, and most organizations underinvest in it.
The distinction matters because the person who builds the model is rarely the person who sustains it. Building requires deep focus over a concentrated period. Sustaining requires distributed attention over an indefinite period. Building is a project. Sustaining is a practice.
And organizations are much better at funding projects than practices.
Four Roles, Not One
The most common pattern for how shared vocabularies die is also the simplest: one person built it, and when that person moved on, nobody picked it up.
Sustainable shared vocabulary requires four roles working together. Not four full-time people — four functions that may be spread across a handful of people or even combined in one or two, depending on the organization's size.
The curator maintains the structure. They ensure the model is internally consistent, properly versioned, well-documented, and up to date. When someone proposes a change, the curator assesses whether it fits the existing structure or would break something. They manage versions, document changes, and notify users. This role requires architectural judgment — the ability to see how individual elements relate to the whole.
The domain expert provides the substance. The curator knows how to structure a model. The domain expert knows whether the model is right — whether the capabilities named actually reflect reality, whether the descriptions match how the work is done, whether a proposed addition fills a genuine gap or duplicates something that already exists. Domain experts typically contribute in focused sessions rather than continuous governance.
The facilitator keeps the model connected to the people who use it. They organize the workshops where leadership teams apply the vocabulary. They collect stories of how the shared language improved a decision. They feed practitioner concerns back to the curator. A model that is technically perfect but not used is worthless. The facilitator's job is adoption.
The sponsor provides authority. Someone with organizational standing must care about the shared vocabulary problem and be willing to fund the modest ongoing investment required to maintain it. Without a sponsor, the vocabulary initiative is a side project maintained by enthusiasts — vulnerable to the first budget cut.
The failure mode is predictable: an organization asks one person (usually an architect or analyst) to do all four roles. They produce a technically sound artifact with no organizational adoption. Then, when they leave, nobody maintains it. The vocabulary decays. And the next time the organization needs shared language, they start from a blank page again.
The Chicken-and-Egg That Stops Everyone
Every shared vocabulary initiative faces a paradox. You need the vocabulary to guide the work of building it. But you need to do the building work to produce the vocabulary. You can't start because you don't have what you're trying to create.
This paradox is real, and it stops organizations before they begin. The perceived scope is overwhelming. "We need to model everything" becomes a multi-year project that requires resources nobody is willing to commit to an unproven concept.
Three strategies break the cycle.
Start with what exists. Most organizations already have shared vocabulary — they just don't call it that. Program classifications, financial reporting structures, org charts, process documentation, strategic plans — these all contain embedded models of how the organization works. Start there. Formalize what's implicit. Don't build from scratch when you can organize what's already known.
Build incrementally. Don't model everything at once. Start with the domain that's most urgent, most visible, or most painful. A partial vocabulary that covers the organization's most strategically important domain is immediately useful. It doesn't need to be complete to be valuable.
Demonstrate value before seeking scale. The biggest risk is premature scaling: asking for broad organizational commitment before proving the concept works. Instead, use the vocabulary to support one governance decision, one strategic conversation, one capability assessment. Show the result. Let the people in the room experience the difference between debating in private languages and debating with shared terms. That experience builds the case for maintaining the vocabulary better than any business case document ever could.
What Keeps It Alive
Once the vocabulary exists, what keeps it alive is governance — but not the heavy, ceremonial kind. The lightest governance that works.
Regular review. Someone looks at the model at a predictable cadence and asks: does this still reflect reality? Monthly if the organization is actively using the vocabulary in governance. Quarterly at minimum. Annual reviews are too infrequent — by the time the review happens, the model has drifted.
Clear ownership. Everyone must know who to contact when a change is needed, who approves structural modifications, and who maintains the documentation. Ambiguous ownership is the same as no ownership.
Version management. When the model changes, stakeholders need to know what changed and why. Otherwise, different teams work from different versions without realizing it — which recreates exactly the vocabulary fragmentation the model was supposed to eliminate.
Decision integration. The vocabulary must be used in actual governance: board papers, investment proposals, project charters, strategy reviews. If the model exists in a repository but is never referenced in the forums where decisions are made, it will atrophy. The link to real decisions is what keeps it alive.
Stakeholder communication. When changes are made, users need to know. Not a forty-page change log. A concise summary: what changed, why, and what it means for how you use the model.
Here's the test that separates living vocabulary from expensive wallpaper: would the leadership team notice if the model disappeared? If the answer is yes — if decisions would be harder, conversations would be slower, alignment would degrade — then the vocabulary is alive and governance is working. If the answer is no, the problem isn't governance. It's relevance. The vocabulary was never connected to decisions that matter.
Four Conditions for Survival
Not every organization is ready to sustain shared vocabulary. Attempting it without the right conditions wastes effort and builds resistance that makes future attempts harder.
Executive sponsorship. Someone with authority must care about the shared vocabulary problem. Not deeply understand reference models — care about the cost of inconsistent language and be willing to fund a modest ongoing solution.
A constituency that feels the pain. Somewhere in the organization, people are spending weeks at the start of every initiative re-establishing definitions. They're watching good ideas die because nobody could agree on what the terms meant. They're seeing the same gaps identified and re-identified across successive projects. They don't need to know the phrase "reference model." They need to feel the problem that a shared vocabulary would solve.
Willingness to invest in shared infrastructure. Shared vocabulary benefits the whole organization but isn't owned by any single function. Like enterprise systems or financial reporting standards, it requires ongoing investment that can't be justified by any single project. Organizations that struggle to fund shared infrastructure will struggle to sustain shared vocabulary.
A culture of collaboration over autonomy. Shared vocabulary works where leaders are willing to adopt common terms even if it means adjusting their local language. It struggles where every unit insists on its own frameworks, its own definitions, its own way of describing what it does.
No organization scores perfectly on all four. The question is which missing condition poses the greatest risk — and whether it can be addressed.
The Morning After Is the Real Test
Every organization can build momentum. The crisis focuses attention. The workshop generates energy. The strategy session produces alignment. Those moments are real, and they matter.
But the test of organizational capability isn't whether you can build shared understanding. It's whether you can maintain it when the crisis passes, when the workshop energy fades, when the strategy session becomes a memory, and when competing priorities fight for the attention and resources that maintenance requires.
The organizations that sustain their shared vocabulary are the ones that treat it the way they treat financial reporting: as foundational infrastructure that everyone uses, not a specialist tool that one team maintains. They invest in it modestly but persistently. They connect it to the decisions that matter. They govern it lightly but consistently.
The organizations that don't sustain it go through cycles: build the vocabulary, watch it decay, start from scratch when the next crisis arrives. Each cycle costs weeks of definitional work that could have been avoided. Each cycle consumes energy that could have been spent on the work that actually matters.
The vocabulary isn't the strategy. But without it, every strategy conversation starts over. And the organizations that tire of starting over are the ones that decide, finally, that maintaining the shared language is worth the investment.
This post is drawn from Building the Common Language, a self-paced course that shows how shared vocabulary is built, governed, and sustained — and why the organizations that maintain their common language execute faster than those that rebuild it every time.