Each day we process a variety of data. Personal data and company data. It is often impossible to see what data it contains from the outside of a document. This makes the extra protection of confidential information a lot more complicated. Fortunately, Microsoft has several intelligent solutions for this.
With data classification, documents are checked for the information they contain. Before this is possible, one must first classify the different types of sensitive information. Microsoft has various pre-trained classifications that you can quickly deploy. Examples are CVs and contracts, but also threats. If these classifications do not (entirely) satisfy your organisation, then you can also train the classifications too. For this, one needs 50 to 500 examples that meet the requirements. Next, you can set up the classification with these examples. To check that the classification works, Microsoft recommends checking 10,000 files as a test.
These files are different. Some comply with the classification, and others do not. If you like the test results, you can publish the classification and put it to use. Via the content explorer, it is possible to get a good idea of the information in Exchange, SharePoint, Teams and OneDrive. But how do you make it visible to the organisation?
Information protection (Labels)
When the data classifications have been set up and possibly trained and tested, the next step is to actually use them in the form of labels. In a label policy, one establishes which sensitive data must be searched for. The document automatically receives the corresponding label when this data is present in a document. Thus, the confidentiality of classified documents becomes visible to everyone in the organisation. The various labels can also be linked to special authorisations or additional security measures so that, for example, they cannot be shared outside the organisation.
Once the labels have been set up, they are published and are ready to be used in practice. Please note that if automatic labelling is used, it must be adequately described with regard to privacy. Would you like to know more about this? Please feel free to contact us.
Need help with the set-up?
We can imagine that it is not entirely clear yet. And that terms such as data classification and labels are not a daily occurrence and do not ring a bell. However, it is essential to start working with them. After reading this blog, do you still have no idea where you should start or how you should set it up? No problem at all. We are happy to help you! Download our product sheet here or contact us.
A while ago, we also wrote an article about the importance of data classification, which you can read here.