Test Data Management

Enterprise Test Data Management

Give development and QA teams the data they need. Without the privacy risk.

Development and testing teams need production-equivalent data to build and test effectively. Providing it safely requires more than copying a production database — it requires masking PII, subsetting to relevant volumes, maintaining referential integrity, and provisioning environments efficiently. DBsafe's test data management capabilities address all of these requirements, turning production-equivalent data into a development asset rather than a compliance liability.

60%
of organisations use real production data in at least one non-production environment
GDPR
developers accessing real customer data without business necessity violates data minimisation
PCI DSS
Requirement 6 prohibits use of live PANs in testing and development
10x
reduction in data breach risk when test environments use properly masked data
The Challenge

Test environments with real data are a persistent compliance liability

The path of least resistance for providing test data is copying a production database. It is fast, it produces realistic data, and it requires no extra tooling. It is also a GDPR violation: personal data in a test environment has no legal basis for processing unless the developers are authorised users with a legitimate need to access that specific customer's data — which they are not.

Organisations that discover real PII in their development environments during a compliance audit face immediate remediation requirements, potential breach notification obligations for anyone who accessed the data, and the cost of rebuilding test environments with properly managed data.

The cost of inaction

Real customer PII in development environments: GDPR Article 5 data minimisation violation

Developer breach of production copy in test environment: full GDPR breach notification obligation

PCI DSS: real PANs in development environment expands CDE scope to include all developer workstations

HIPAA: real PHI in test environment requires same safeguards as production — rarely applied

The DBsafe Approach

How DBsafe manages test data safely

01
01

Discover and configure masking rules

DBsafe scans your production databases and automatically suggests masking rules for every sensitive column. Rules are reviewed and activated once, then reused for every test data refresh.

02
02

Create masked, subsetted test datasets

DBsafe creates test datasets by masking sensitive data and optionally subsetting to a relevant volume. Referential integrity is preserved across all related tables. The process runs in minutes for typical databases.

03
03

Provision environments on demand

Via API or CI/CD integration, trigger fresh test data creation whenever a new environment is needed. Development teams get current, representative, safe data without waiting for manual data preparation.

Why test data management matters for compliance

GDPR's data minimisation principle requires that personal data be processed only to the extent necessary for the specified purpose. Development and testing does not require access to real customer data — it requires data that behaves like real customer data. This distinction is the foundation of the compliance argument for test data management: replace real data with realistic substitutes, and you eliminate the legal basis problem.

Beyond GDPR, PCI DSS explicitly prohibits the use of live PANs in testing and development. HIPAA requires the same safeguards in development environments that apply to production when real PHI is present. Test data management solves these issues across all frameworks simultaneously.

The four types of test data

Test data management addresses four types of data needs. Production-equivalent masked data: real production schema and data volumes with all sensitive values replaced. This is the most realistic and the most resource-intensive to produce. Subsetted masked data: a representative subset of production data, masked. Appropriate when full production volumes are not needed for testing. Synthetic data: generated data that matches the statistical properties of production. Slower to produce but has no production data dependency. Golden datasets: hand-crafted datasets for specific test scenarios.

DBsafe supports all four types, enabling test teams to choose the right approach for each testing context.

Subsetting and relationship preservation

A production database may contain hundreds of millions of rows across hundreds of tables. Creating a full masked copy for every test environment is impractical. Subsetting reduces this to a manageable volume — typically 1-5% of production — while preserving all the relationships that make the data representative.

DBsafe's subsetting engine starts from a set of root entities (e.g., the last 10,000 customers) and follows foreign key relationships to include all related records: their orders, transactions, communications, preferences, and audit history. The result is a consistent, internally coherent subset that tests all application code paths.

Referential integrity in masked test datasets

Referential integrity is the most common masking failure. When an email address in the customers table is masked to user123@example.com but the same email in the order confirmation emails table is masked to user456@example.com, the relationship is broken and tests that join these tables will fail.

DBsafe tracks every occurrence of each sensitive value and ensures that the same masked substitute is used consistently across all tables. The masked test dataset is referentially complete: every join that works in production works identically in the masked copy.

Refreshing test environments at DevOps speed

The value of masked test data depends on it being current. A test environment refreshed three months ago may not reflect the current production schema or data patterns. DBsafe's incremental masking capability enables frequent refreshes: only rows that have changed since the last mask are reprocessed, making nightly or even hourly refreshes practical for most database sizes.

Refreshes are triggered via API, enabling integration with CI/CD pipelines, GitOps workflows, and scheduled tasks. Development teams can request a fresh environment as easily as triggering a build.

Compliance evidence for test data management

Having a test data management programme is one thing; proving it is working is another. DBsafe generates masking attestation reports for every test data refresh: which columns were masked, which technique was applied, whether integrity checks passed, and when the operation occurred.

These reports satisfy auditor requests to demonstrate that non-production environments contain no real personal data. They can be referenced in compliance evidence packages for GDPR, PCI DSS, and HIPAA, supporting scope reduction claims for non-production systems.

Platform Capabilities

Everything you need to address test data management — in one integrated platform.

Production-Equivalent Masking

Mask every sensitive column while preserving the realistic feel of production data.

Intelligent Subsetting

Create representative subsets following foreign key relationships from seed entities.

Referential Integrity

Consistent masking across all related tables ensures every join works in the masked dataset.

API-Driven Provisioning

Trigger test data creation from CI/CD pipelines and environment provisioning workflows.

Incremental Refresh

Refresh only changed data for fast, frequent test environment updates.

Masking Attestation

Per-refresh compliance evidence documenting masking coverage and verification results.

Frequently Asked Questions

Answers to the most common questions about test data management.

What is test data management?

Test data management is the practice of providing development and testing teams with production-equivalent data that has had all sensitive personal information removed or replaced. It includes masking PII, subsetting data to manageable volumes, preserving referential integrity, and provisioning environments efficiently and repeatably.

Is copying production data to development environments illegal under GDPR?

Copying production personal data to development environments without a lawful basis and without equivalent access controls is likely to violate GDPR's data minimisation and purpose limitation principles. The ICO has indicated that personal data used in development must be appropriately controlled. Masking before providing data to developers eliminates the compliance issue.

How does subsetting work in DBsafe?

Subsetting starts from a defined set of root entities — typically the most recent N customers, orders, or records of a specific type. DBsafe follows all foreign key relationships from these root entities and includes every related record across all tables. The result is a complete, internally consistent subset that tests the same application code paths as the full dataset.

Can DBsafe refresh test environments automatically on a schedule?

Yes. DBsafe supports scheduled masking jobs that refresh test environments on a configurable schedule. Daily or nightly refreshes can be configured to run automatically, with completion notifications sent to development team channels. Incremental masking reduces refresh time for regular schedules.

Does DBsafe generate compliance evidence for test data masking?

Yes. DBsafe generates masking attestation reports for every masking job. These reports document which columns were masked, which technique was applied, when the masking ran, whether post-masking verification found any unmasked sensitive data, and who triggered the job. These reports support scope reduction claims for non-production environments in PCI DSS and HIPAA assessments.

Ready in 5 minutes

Give your development team production-quality data — safely

DBsafe creates masked, referentially intact test datasets that your team can use freely, with compliance evidence ready for every audit.