Prepare and test your skills

Prepare and test your skills

Establishing trust through data quality and consistency means using data governance policies to keep data accurate, reliable, and secure so stakeholders can make informed decisions. Data governance is the set of policies, procedures, and technology that keeps data secure, private, accurate, available, and usable. In Data Studio (opens in a new tab) (formerly Looker Studio), governance controls who can access, edit, and share reports. These controls apply to both the no-cost version of Data Studio and Data Studio Pro, and they support data democratization by giving more people access to data while still protecting sensitive information.
Why data governance builds trust is that it protects sensitive information, ensures data integrity, meets compliance requirements, and maintains stakeholder confidence. Data Studio reports often contain personally identifiable information (PII), financial data, and proprietary business strategies. Controlling who can access this data prevents breaches and unauthorized disclosure, while restricting modification rights to authorized personnel keeps reports accurate for decision-making. Compliance features help organizations follow regulations such as GDPR and HIPAA, which reduces legal risk and shows clients and partners that data is handled responsibly.
Data credentials in Data Studio determine whose authorization is used to show data from a data source. The choice changes whether viewers need their own access to the underlying dataset and how much control the owner keeps.
| Credential type | How data access works | Best for |
|---|---|---|
| Owner's Credentials | Other users view data using the credential owner's authorization and do not need their own dataset access | Convenience when viewers lack direct access |
| Viewer's Credentials | Each viewer uses their own credentials, so only people with dataset access can see data | Maximum control over who sees data |
| Service Account Credentials | A non-human Google account is authenticated and authorized to access data | Automated or application access with controlled permissions |
For maximum data access control, use Viewer's Credentials or Service Account Credentials when possible, because Owner's Credentials lets viewers see data without their own dataset access.
Exam tip: For maximum data access control, use Viewer's Credentials or Service Account Credentials (if possible) for your data sources.
Reusable versus embedded data sources in Data Studio differ in how they are created and who can edit or copy them. An embedded data source is created while editing a report, and anyone who can edit that report can also edit the embedded source and its connection. A reusable data source is created from the Data Studio home page, and only users with explicit access can edit it. Sharing or copying a report with an embedded source also shares or copies the source, so viewers may gain access to data and metadata by copying the report. For maximum control, use reusable data sources in reports and convert an embedded source to reusable when needed.
Hiding sensitive fields and limiting sharing in Data Studio reduces exposure of confidential data while keeping the rest available for analysis. If a dataset includes fields that should not appear in reports, a data source editor can hide a field to remove it from the report field list or remove it from the data source entirely. Hidden fields remain visible to data source editors who can unhide them, while removed fields are taken out completely. Governance also limits how reports are shared by restricting sharing to specific people or groups, preventing downloading, printing, and copying, and preventing editors from changing access. Row-level security goes further by letting different viewers see only the rows they are permitted to see.
Governance for AI and machine learning (ML) ensures models are built, deployed, and used responsibly in a secure environment. Gemini Enterprise Agent Platform provides a governed environment for developing and deploying ML models. Features such as model explainability and fairness metrics help address responsible-AI concerns, while alignment with the Secure AI Framework (SAIF) (opens in a new tab) provides a practical approach to AI security and risk. Tracking accuracy, false positives in fraud detection and security alerts, and efficiency gains from automation helps stakeholders trust ML outcomes because model quality and fairness are visible.
Governance for compliance and sensitive data uses classification and boundary controls to identify and protect regulated information. Sensitive Data Protection identifies, classifies, and protects sensitive data, including financial information. Assured Workloads sets up data boundaries in Google Cloud for sensitive and regulated workloads and helps meet requirements such as FedRAMP and CJIS, including data residency, audit trails, and data protection standards. Together these controls help ensure data stays accurate, consistent, and trustworthy for business decisions and ML outcomes.
Mitigating risk through security and regulatory compliance means building security controls and legal requirements directly into data governance across the data lifecycle. Data governance creates the framework for data integrity, security, and accountability from collection through storage and use. Security controls protect sensitive information from unauthorized access, while compliance standards prove that protection meets legal and industry expectations. Together they reduce the risk of breaches, misuse, and non-compliance throughout the data journey.
Compliance standards within governance are recognized benchmarks that define how sensitive data must be protected and reported. A governance strategy aligns its controls and monitoring with regulatory compliance (opens in a new tab) frameworks including:
Security Command Center helps enforce this alignment by assessing and reporting security benchmark compliance. Data profile findings can flow from discovery into Security Command Center, so compliance reporting reflects the actual sensitive data found. This connection keeps governance accountable because the same discovery results guide both protection actions and compliance evidence.
Security controls for sensitive information are technical safeguards that limit who can see data and transform data to reduce exposure. Sensitive Data Protection provides discovery, inspection, and de-identification tools, while Identity and Access Management (IAM), column-level security, and row-level security hide or display data based on who is viewing it. For data in motion, the Cloud Data Loss Prevention API accepts data from any workload for near real-time inspection and masking before that data reaches backend storage. A common in-motion protection flow moves data from the application to safe storage in order:
content.deidentify to detect and transform sensitive elements in the string.This sequence prevents raw sensitive data from entering storage because protection happens before persistence. Choose in-motion de-identification when sensitive data might arrive through user input, and choose IAM or column-level and row-level security when sensitive data must remain in storage but be visible only to authorized viewers.
Discovery, monitoring, and investigation across the lifecycle show where sensitive data lives, whether it stays protected, and how large the risk is. Discovery runs at the organization, folder, or project level to identify what kind of data is stored. Continuous monitoring watches resources already known to hold sensitive data, and inspection jobs investigate resources where further analysis is needed.
| Challenge | Action |
|---|---|
| Need to know what kind of data is stored | Run discovery at the organization, folder, or project level |
| Sensitive data found in an already protected resource | Continuously monitor the resource by running discovery and automatically exporting profiles to Security Command Center, Google SecOps, and Knowledge Catalog |
| Sensitive data found in an unprotected resource | Hide or display data based on viewer using IAM, column-level or row-level security, and transform or remove sensitive elements with Sensitive Data Protection de-identification |
| Need to understand the extent of data risk | Run an inspection job on the resource, and use synchronous content methods of the DLP API to proactively block sensitive data from entering storage |
Exam tip: Use discovery to learn what kinds of data you store, and use an inspection job to investigate the extent of risk after sensitive data is found.
Enhancing data discovery and transparency via metadata management means describing what data exists and showing its origin, transformations, and movement. In Google Cloud, Knowledge Catalog (opens in a new tab) provides metadata management, search, and lineage so users can find trusted assets and understand their history. This visibility improves data integrity, accountability, and confidence across the data journey.
How metadata management organizes data knowledge in Knowledge Catalog is by creating consistent descriptive information about assets such as tables, files, and dashboards. Metadata describes what an asset contains, who owns it, and how it should be used. Because the catalog uses a flexible metamodel, the same system can describe many different kinds of assets in a consistent way. The metamodel organizes this information into related building blocks:
Together these blocks let an organization keep metadata consistent while covering diverse systems and use cases.
How data lineage shows origin and movement in Knowledge Catalog is through an end-to-end record from source to destination. Lineage visualizes transformations at the column level, so users see not only which tables were involved but how individual fields flowed and changed. Data flows from source systems through transformation steps to destination tables, reports, and dashboards, and lineage preserves that path. This transparency helps users judge whether data is fit for use and understand downstream impact before making a change. If a source column changes, lineage shows which downstream assets depend on that column and could be affected.
A process flow showing column-level data lineage: columns flow from a source table through transformation steps to destination tables, reports, and dashboards, with Knowledge Catalog recording the path so users can see where each column came from and which downstream assets depend on it.
How discovery and enriched metadata help users find trusted data is by making search rely on cataloged context instead of tribal knowledge. Asset discovery lets users search across the organization to quickly find existing tables and files and understand their context. This reduces duplication and helps teams reuse trusted data. Enriched metadata adds business and technical context such as descriptions, classifications, and glossary terms, and teams can manage aspects to keep entries complete. An AI enrichment agent built with the Agent Development Kit can automate metadata curation and glossary mapping to keep that context up to date.
Exam tip: Metadata describes what a data asset is, while data lineage shows where the asset came from, how it was transformed, and where it goes next.
Owner's Credentials lets other users view data using the credential owner's authorization without needing their own dataset access, while Viewer's Credentials requires each viewer to use their own credentials so only people with dataset access can see data, and Service Account Credentials uses a non-human Google account authenticated and authorized for automated or application access. For maximum data access control, use Viewer's Credentials or Service Account Credentials when possible because Owner's Credentials lets viewers see data without their own dataset access.
Use a reusable data source in Data Studio for maximum control because it is created from the Data Studio home page and only users with explicit access can edit it. An embedded data source is created while editing a report and anyone who can edit that report can also edit the embedded source and its connection, and sharing or copying a report with an embedded source also shares or copies the source so viewers may gain access to data and metadata.
The Cloud Data Loss Prevention API protects data in motion when the application accepts user-provided input, sends it to the API for content.deidentify to detect and transform sensitive elements, and then saves only the de-identified string to backend storage instead of the original string. This sequence prevents raw sensitive data from entering storage because protection happens before persistence.
A financial services company is designing a data governance strategy on Google Cloud to mitigate regulatory risks across its data lifecycle. To comply with strict privacy regulations such as GDPR and CCPA, the company must meet the following requirements:
Which Google Cloud service should the company integrate into its governance framework?
Last updated