Data Standards and Definitions: The Common Language Problem

Opening Scene

In a busy medieval marketplace, a “bushel” from one region’s merchant holds a different volume than a “bushel” from the next town over, and every transaction between them requires a slow, error-prone negotiation over what the word actually means — until a common standard of weights and measures, imposed and enforced across the whole kingdom, lets any two merchants trade without first arguing about vocabulary.

In Plain English

A data standard defines, precisely and in one place, what a term like “active customer” or “monthly revenue” actually means, so that every team calculating it produces the same number. Without shared definitions, two perfectly correct calculations can still disagree, not because either is wrong, but because they’re quietly answering slightly different questions.

The Old Way

Before shared definitions were treated as governance infrastructure:

Every department maintained its own private definition of common business terms, discovered to be different only when someone compared two reports side by side.
Metric definitions lived inside SQL queries or spreadsheet formulas that nobody outside the original author could easily inspect or question.
Disputes about “whose number is right” consumed meeting after meeting, without anyone realizing the real problem was a definitional mismatch, not a calculation error.

A data standard resolves that dispute before it starts, by putting the definition itself, not the calculation, at the center of the disagreement.

What’s Changing (and Why AI Is the Reason)

Organizations increasingly maintain a centralized, searchable business glossary rather than letting definitions live privately inside individual reports.
This connects to this content library’s dedicated data cataloging and lineage series, since a catalog is what makes a standard definition discoverable and traceable back to the systems where it’s actually applied.
AI systems make definitional drift far costlier than before, since a model trained on inconsistently defined data will produce confidently wrong outputs that are much harder to trace back to the root definitional mismatch than a wrong number in a spreadsheet.

The Metaphor, Fully Extended

The Merchant’s Bushel	The Business Metric
Different regions measuring the same word differently	Different departments defining the same term differently
Slow, error-prone haggling before every trade	Repeated disputes before every cross-team report
A kingdom-wide standard of weights and measures	An organization-wide business glossary
Any two merchants trading without vocabulary disputes	Any two teams reporting the same number without disputes

For Beginners: What to Actually Do

Before trusting a number, check whether the term behind it has a documented, agreed-upon definition anywhere.
When you spot two conflicting numbers for the same metric, check the definitions before assuming a calculation error.
Contribute definitions for terms you own to a shared glossary if your organization has one, or propose starting one if it doesn’t.

For Practitioners and Leaders: The Deeper Layer

Centralize business term definitions in a governed glossary tied to the data catalog, not scattered across individual teams’ documentation.
Assign a single accountable owner to each contested term, with the governance council as the escalation path when departments disagree.
Audit high-visibility metrics periodically for definitional drift, since even a governed term can quietly diverge as systems and processes change.

Quick Recap

Data standards define common terms precisely so everyone calculating them gets the same answer.
Definitional mismatches, not calculation errors, are often the real cause of conflicting numbers.
Centralized, searchable glossaries are replacing definitions buried in private reports and queries.
AI makes the cost of definitional drift significantly higher than it used to be.

Where This Fits in the Series

Article 7 covered writing policies people actually follow. Article 8 addressed the deeper common-language problem those policies depend on: agreeing what terms actually mean. Article 9 turns to a related accountability question — not what things mean, but who’s responsible when something built on those definitions breaks.