Picture a textile company where the accounting manager pulls the monthly sales figures from the finance system. On the same day, the sales manager runs the same period’s numbers from the sales tracking program. When both reports land on the same table, the numbers do not match. Accounting says ‘net sales,’ sales says ‘gross revenue.’ Both are right by their own definitions. Both are reading exactly what their program shows them. But in the management meeting, which number does the owner believe? This is the real problem sitting at the center of most data warehouse efforts.
A data warehouse — a central system that pulls records from different programs into one place for reporting — sounds like a straightforward technical project. The server gets set up, connections between programs are established, and data flows in from multiple sources. But when the reports still come out inconsistent, everyone points at someone else. The IT person says the data was transferred correctly. The accountant says the records are accurate. The sales manager says the same. What is missing is not a technical fix. Nobody has decided who owns which data, what each term means across the company, or who is responsible when a number is wrong. This gap is what data governance addresses.
Data governance is a simple idea dressed up in complicated language. Every piece of data needs an owner. Think of a warehouse with no stock manager — nobody knows what is where, counts never match, and every inventory check turns into an argument. Data works the same way. Who decides what the word ‘customer’ means in this company? Does it include everyone who received a quote, or only those who were invoiced? The answer will differ from one business to another. But inside a single company, every department must use the same answer. The person who sets and defends that answer is the data owner for customer records.
In practice, this means assigning a responsible person for each main category of data. The sales manager owns customer data. The production chief owns product and materials data. The accounting manager owns financial definitions. These people do not need to be technical. Their job is to write down what their data means, check that it stays accurate, and make sure updates happen when something changes. When the accounting manager writes and signs off on the definition of ‘net sales,’ that definition becomes the company standard. Nobody else gets to use a different version. Simple on paper — but it takes real organizational discipline to hold the line.
The most visible benefit is that management reports become trustworthy. When two different numbers show up in the same meeting, the discussion stalls. Hours go into defending each figure, and decisions get delayed. Once definitions and ownership are locked in, that argument disappears. Everyone sees the same ‘net sales’ figure because everyone is working from the same definition. A second benefit shows up in training. When data definitions are written down — which fields are required when opening a customer record, what counts as a completed order — new staff can learn faster. Mistakes from misunderstanding drop off quickly.
The hardest part of data governance is not technical. It is human. Giving someone data ownership also means giving them accountability. The accounting manager now answers for the accuracy of financial figures. Some managers resist this. Others fight over the same data — ‘is the customer record owned by sales or accounting?’ These turf disputes can drag on for months. If senior management does not step in and make a decision, the whole effort stalls. The system gets built, the owners get named on paper, but nobody enforces the rules. Within a year, the old confusion is back.
For small and mid-sized businesses considering a data warehouse, or struggling to understand why existing reports never agree, the first question to ask is this: does every major category of data in this company have a written definition, and is there a named person responsible for it? If the answer is no, no amount of technical investment will produce reliable reports. The real foundation is laid not in the server room but in the meeting room — sitting down with department heads, agreeing on definitions, and assigning responsibility before a single cable is connected.
A KOBİ owner who wants consistent numbers does not need to become a technology expert. The practical step is to organize a working session with key managers, list the main data categories the business runs on, and ask each manager to write one paragraph defining what their data means and what makes it accurate. That document, however rough, is the start of data governance. Everything technical built on top of it will be more reliable for having it.
This article was originally written in Turkish by Gökhan MERCANOĞLU on January 26, 2004 and has been automatically translated into English and other languages using machine translation.