Domain 1 of 3

Implement information protection

Domain · 30–35% of the SC-401 exam

Know your data, then protect it: recognizing content never protects it

Take a requirement such as: find every document that holds a customer account number, and let only the finance team open it. Microsoft Purview Information Protection gives you tools to know your data, protect your data, and prevent data loss; this domain covers the first two, and preventing data loss is covered in Create and configure data loss prevention policies. To know your data, a classifier recognizes content: sensitive information types (SITs) are pattern-based classifiers, and machine learning identifies items through trainable classifiers. Within SITs, this guide's narrower pattern-based SIT means built-in, named entity (SITs that detect person names, physical addresses, and medical terms and conditions), credential and custom SITs matching a regular expression, keyword list, or function, beside exact data match (EDM) SITs and fingerprint SITs created from a standard form. To protect your data, capabilities led by sensitivity labels apply protection actions that include encryption, access restrictions, and visual markings; encryption applied by sensitivity labels to documents, emails, and meeting invites uses the Azure Rights Management service, except when using S/MIME for Outlook (S/MIME: a protocol for sending signed and encrypted messages), and Microsoft Purview Message Encryption is built on Azure Rights Management too. A classifier only identifies content, and a sensitivity label you create stays a definition until it is applied and apps and services enforce its settings.

The domain unfolds in three steps: one to know your data, two to protect it

Know your data is Implement and manage data classification: translating requirements into pattern-based SITs, fingerprint SITs, EDM SITs or trainable classifiers, turning on optical character recognition (OCR) so that you can extend classification to images without creating separate classifiers, and checking what is classified in Data explorer and Content explorer. Protect your data: sensitivity labels is Implement and manage sensitivity labels in Microsoft Purview: choosing each label's scope (items, or containers such as Teams, Microsoft 365 groups and SharePoint sites), configuring encryption and content markings (the visual markings named above), and delivering labels through publishing policies and auto-labeling policies, including labels that extend to Power BI and, through Microsoft Defender for Cloud Apps, to content in third-party apps and services. Protect your data: Windows, file shares, and Exchange is Implement information protection for Windows, file shares, and Exchange: the Microsoft Purview Information Protection client, which runs on Windows only, replaces the Azure Information Protection (AIP) unified labeling client, and extends sensitivity labels beyond labels that are built into Microsoft 365 apps and services (despite the shared name, it is one client, not Microsoft Purview Information Protection as a whole); its information protection scanner, which discovers, labels, and encrypts files on network shares and SharePoint Server libraries; and Microsoft Purview Message Encryption, which encrypts email messages and attached documents sent to any user on any device and which an administrator can also apply with Exchange mail flow rules (also known as transport rules), with Microsoft Purview Advanced Message Encryption adding expiration and revocation for email sent to external recipients.

Report first, then enforce

Sensitivity labels, auto-labeling policies and the scanner each offer a stage before full protection. An auto-labeling policy's simulation and the scanner's discovery mode only report, while two stages apply real labels: a label with no protection settings, which can generate usage reports and activity data, so you can choose to apply protection settings later, and a publishing policy used to pilot labels with just a few users before a policy for all users. Because users don't interact with service-side labeling (an auto-labeling policy labeling content already saved (in SharePoint or OneDrive) or emailed (processed by Exchange Online), not Office apps during editing), Microsoft advises that you run an auto-labeling policy in simulation before you turn it on. The scanner setup has you create the content scan job (the scanner's settings for which repositories it scans and how) with Enforce sensitivity labeling policy set to Off and run the first scan in discovery mode, then change Enforce to On once the reports look right.

Know your data and protect your data: the step each subtopic covers

StepWhat it decidesKey controlsDrill into
Know your dataWhat the content is, and which items are classifiedPattern-based SITs, fingerprint SITs, EDM SITs, trainable classifiers, OCR, Data explorer and Content explorerImplement and manage data classification
Protect your data: sensitivity labelsHow labeled items and containers are handled, and who can open encrypted contentLabel scopes, encryption and content markings, publishing policies, auto-labeling policies, container labels, labels in Power BI and Defender for Cloud AppsImplement and manage sensitivity labels in Microsoft Purview
Protect your data: Windows, file shares, and ExchangeHow labels reach files on Windows outside Office apps, or in on-premises repositories, and how email is encryptedInformation Protection client, information protection scanner, Message Encryption, Advanced Message EncryptionImplement information protection for Windows, file shares, and Exchange

Decision tree

Only recognizethe content?YesSIT or trainableclassifierNoExpire or revokeexternal mail?YesAdvanced MessageEncryptionNoMail flow ruleencrypts email?YesMessageEncryptionNoLabel on Windowsoutside Officeapps, or inon-premisesrepositories?YesInformationProtection client,or its scannerfor repositoriesNoSensitivity label, publishedor applied automatically

Subtopics in this domain