No committee, no 50-page policy. Just the few rules that keep your data usable.
"Data governance" sounds like something for banks and large corporations: committees, data stewards, long policies nobody reads. So most SMEs skip it, and their data slowly drifts until a CRM move, an audit or an AI project brings it to light.
You don't need the corporate version. You need a light one that fits on a page and that people actually follow.
You maintain your machines, your vehicles and your buildings, because they lose value if you don't. Customer and business data works the same way. Contacts leave, companies move, fields get filled in differently by different people. Without upkeep, the data loses value every month, and the systems built on it lose value with it.
Data you look after pays back in faster answers, fewer errors, reports people trust, and a solid base for anything new, from a CRM migration to AI.
Lightweight governance comes down to five questions. Write the answers down, put them where people can find them, and review them once a year.
1. Where does each type of data live? Customers, contacts, deals, products, invoices. One system is the source of truth for each. Others may read from it but don't create their own copies.
2. Who owns it? Not IT, but the people who use it. Account managers own their customer records. Someone in finance owns payment terms. Ownership means you keep it correct and you decide when something is unclear.
3. What are the rules? A short list: which fields are mandatory, how company names are written, which values are allowed for country, industry and lead source. Build the rules into the system as required fields and dropdowns, so following them is the easy option.
4. Who can see and change what? Access by role, not by habit. That protects sensitive data, keeps you in line with the revised Swiss data protection law, and stops well-meant edits in the wrong place.
5. How do we keep it clean? A short review every quarter: duplicates, incomplete records, contacts nobody has spoken to in years. Small and regular beats a big clean-up every five years.
Many governance concepts were built for large organisations with dozens of systems and data teams. Data mesh, data contracts, formal stewardship councils: useful ideas at that scale, but for a company with one CRM, one ERP and a few tools, they add overhead without much benefit. Start with the five questions above. If you grow into the bigger concepts later, you'll have the foundation for them.
Want to know how mature your data handling really is? Take our free AI-Readiness Check: twelve questions, about three minutes, and you see your score in five areas straight away.
If you're moving systems soon, governance is easiest to set up at the same time. Read clean your data before you move it, and see what bad customer master data really costs.
Our Digitalization Audit answers the five questions with you, based on your real systems and processes, and gives you a ranked plan for what to fix first.