Data Quality, Governance & Access

A dashboard is only as good as the data beneath it, and that data is only trustworthy if someone has taken care of it. Data governance is the unglamorous but essential discipline of making sure the right people can reach the right data, that everyone agrees on what each number means, and that someone is genuinely responsible when something looks wrong. It is the part of the hidden plumbing that decides whether you can actually believe the figure on your screen. Get it right and arguments end; get it wrong and every meeting starts with “but whose number is correct?”

Data quality: is the number even right?

Data quality is the foundation everything else rests on. High-quality data is accurate, complete, consistent and current — no duplicate customers, no missing dates, no figure that is three weeks stale. Poor quality is quiet and corrosive: a single duplicated record can inflate a sales total, and a date stored in the wrong format can drop a whole region out of a report without any error message at all. Because bad data rarely announces itself, the people who rely on it often trust it longest. That is why quality has to be actively checked, not assumed.

Shared definitions: same word, same meaning

One of the most common causes of conflicting reports is not bad data but inconsistent definitions. If your team counts a sale when the order is placed and finance counts it when the invoice is paid, both dashboards are “correct” and they will never agree. Governance fixes this by writing down agreed definitions — a shared dictionary that says exactly what “active customer,” “revenue,” or “complete” means across the whole company.

If two reports disagree, the problem is usually a definition, not a calculation. Settle the words first.

Access and permissions

Not everyone should see, or change, everything. Access control is about granting each person the permissions their role needs and no more — a principle you may know as least privilege. Think of a ladder: most people only need to view a report; a smaller group can edit the underlying dataset; a very small group can administer it and grant access to others. This protects sensitive information, like salaries or personal customer details, and it protects the data itself — fewer hands able to change a shared dataset means fewer accidental, hard-to-trace mistakes. When you need access, the right move is to request it formally so your permissions are tied to you and removed when you move on.

Ownership: someone has to be responsible

Trusted data has an owner. A data owner or steward is a named person accountable for a particular dataset — its quality, its definitions, and who can use it. Without an owner, a problem becomes everybody’s job and therefore nobody’s: the figure is wrong, three teams shrug, and it stays wrong. With a clear owner, there is someone to ask, someone to fix it, and someone who cares whether it is right. This single role quietly underpins whether a number can be trusted at all.

Spot it: governance issues

Read each situation and decide for yourself, then tap a card to flip it and check your answer.

Sort the governance concepts

Drag each item into the bucket it belongs to — or tap an item, then tap a bucket. Hit Check placement when you’re done.

How to use it

You can strengthen governance just by behaving like a good citizen of the data. Before you act on a surprising number, check who owns the dataset and what the figure is defined to mean. When you build something in a self-service tool, use the official shared dataset rather than a personal export. Request access through the proper channel instead of borrowing a colleague’s login. And when two reports disagree, ask about definitions before assuming someone made a mistake. Useful phrases: “Who owns this dataset?” or “How exactly are we defining revenue here?” or “Can I get view access to that report the proper way?” Trusted data is not an accident — it is the result of small, deliberate habits like these, repeated by everyone.