Data Governance in Plain English
Imagine you work at a company with 500 employees. Your colleagues in sales have a list of your top clients. Your marketing team has a slightly different list. Finance has yet another version. Each team has been updating theirs independently for the past two years.
Now imagine somene asks: "How many clients do we actually have?" No one will answer because the data isn't governed.
So what exactly is data governance?
Data governance is the set of rules, roles, and processes that decide who is responsible for data, what standards it must meet, and how it can be used across an organisation.
Think of it less like IT infrastructure and more like a legal system for your data. Laws do not build roads — but without laws, roads would be chaos. Governance does not store data, but without it, data becomes unreliable, inconsistent, and ultimately useless.
The three things governance actually does
- Assigns ownership: Someone is responsible for each dataset. They maintain it, validate it, and answer for it.
-
Sets standards: A client is defined the same way everywhere. "Active" means the same thing in sales and in finance.
-
Controls access: The right people can access the right data — and the wrong people cannot.
Why should non-technical people care?
When a sales manager presents a forecast based on duplicate customer records, the strategy that follows is built on sand. When a marketing team targets a segment defined differently from the one finance is tracking, the campaign results will never make sense.
Data governance is, at its core, an organisational discipline. It requires decisions made by humans — not just systems configured by engineers.
What does it look like in practice?
Governance is not a product you install. It is a practice you build. Here are some of its most common building blocks:
A data catalogue — a shared dictionary of what data exists, where it lives, and who owns it. Think of it as an index for your organisation's information assets.
Data owners — specific people (not just teams) accountable for the quality and accuracy of a particular dataset. Without a named owner, nobody is responsible. Which means nobody fixes anything.
Data definitions — agreements on what words mean. What is a "customer"? A lead who filled in a form? Someone who made a purchase? Someone active in the last 12 months? If the definition changes depending on who you ask, the numbers will never align.
Access policies — clear rules about who can read, edit, or delete data, and under what circumstances. This matters for both security and compliance (GDPR, for instance, requires this by law).
Data quality rules — standards that data must meet to be considered usable. A phone number field that accepts text. An email field with no format validation. A date column storing "TBD". These are governance failures masquerading as technical ones.
The human side of governance
Teams resist sharing data because they see it as a competitive advantage within the organisation. Managers resist defining standards because definitions force accountability. Executives deprioritise governance because it is invisible when it works and only visible when it catastrophically fails.
Effective data governance requires sponsorship from leadership, genuine collaboration across departments, and a culture that treats data as a shared organisational asset — not a departmental possession.
That is why the best data governance programmes are led by people who understand both the business and the data. Not just engineers building pipelines, and not just managers issuing mandates. People who can translate between both worlds.