Static Data Masking

Static Data Masking for Safe Development Environments

Create a permanent, production-equivalent database with all sensitive data replaced.

Static data masking creates an irreversibly altered copy of your database where sensitive data has been replaced with realistic substitutes. The result is a masked dataset that looks, feels, and behaves like production data — but contains no real personal information. DBsafe performs static masking with referential integrity preservation, format compatibility, and full audit evidence for compliance reporting.

3.1x
more likely to suffer a breach if production data is used in test environments
60%
of organisations use real production data in non-production environments
PCI DSS
Requirement 6 restricts use of PANs in non-production environments
Zero
real PII remaining after a complete DBsafe static masking job
The Challenge

Why static masking is the right approach for persistent environments

When developers, testers, and analysts need access to production-equivalent data, the options have traditionally been limited: share real production data (creating compliance risk), or build synthetic datasets from scratch (expensive and often not representative). Static masking offers a third way: take production data and replace all sensitive values with realistic substitutes, creating a masked copy that developers can work with freely.

Static masking is appropriate for environments that persist between deployments: shared development databases, QA environments, partner access datasets, and training environments. The masking runs once per refresh, and the result is a complete, self-contained database with no sensitive data.

The cost of inaction

Developer exposure to real customer data without business necessity violates GDPR Article 5

Third-party access to unmasked data without adequate DPA creates joint controller liability

PCI DSS failure if PANs are present in development and testing scope

Training environments with real data breach regulatory requirements in the event of a leak

The DBsafe Approach

How DBsafe performs static masking

01
01

Source database analysis

DBsafe connects to your source database and analyses the schema, identifying sensitive columns via automated discovery. Masking rules are suggested automatically and can be reviewed and customised before the masking job runs.

02
02

Masking job execution

The masking engine processes the source database and writes a masked copy to the target. Transformations preserve referential integrity, format compliance, and data type constraints. Progress is logged in real time.

03
03

Verification and evidence

After masking, DBsafe runs a verification scan confirming that no sensitive data remains in the masked copy. A masking attestation report is generated for compliance evidence, documenting every column masked and the technique applied.

Use cases for static data masking

Static masking is the right choice when you need a complete, persistent masked copy of a database. Development environments that are shared across a team benefit from a single masked database that is refreshed periodically from production. QA environments need production-representative data that does not change between test runs. Data shared with third-party developers, auditors, or outsourcing partners must be fully masked before transmission.

Static masking is also appropriate for data archival: older backups that are retained for regulatory or operational reasons but no longer need to contain real personal data can be statically masked, reducing the compliance scope of your archival systems.

Masking techniques used in static masking

DBsafe supports multiple techniques within a single static masking job. Different columns can use different approaches based on the data type and the use case. Email addresses are typically pseudonymised to preserve uniqueness and format. Names are replaced with random plausible substitutes. Dates of birth are shifted within a range to preserve age distribution without revealing real birthdates. Credit card numbers are replaced with Luhn-valid dummy values.

The selection of technique is guided by two factors: the need to preserve data utility for the target use case, and the requirement to make the original value unrecoverable. DBsafe's configuration interface makes these trade-offs explicit.

Referential integrity in static masking

A masked database must be functionally equivalent to its source for development purposes. This means that joins must work correctly across all tables. If a customer email address is masked in the customers table, all references to that email in the orders, communications, preferences, and audit tables must receive the same masked value.

DBsafe analyses foreign key relationships before running a masking job and builds a consistent masking map. Every occurrence of a sensitive value across all tables is traced and masked consistently. The result is a masked database where every join still works and every application function operates as expected.

Performance and large-scale masking

Enterprise databases often contain hundreds of millions of rows. DBsafe's masking engine is built for scale: it uses batch processing, parallel execution, and streaming writes to handle large datasets without excessive memory consumption or extended downtime.

Masking jobs can be configured with concurrency settings, chunk sizes, and priority levels. For very large databases, incremental masking — masking only changed rows since the last run — reduces the time required for periodic refresh cycles.

Static masking and compliance scope reduction

One of the most valuable properties of static masking is its ability to reduce compliance scope. A properly masked development environment may be excluded from PCI DSS cardholder data environment scope, reducing the cost and effort of annual assessments. HIPAA safe harbour de-identification, applied via DBsafe's masking rules, may exclude development environments from covered entity obligations.

DBsafe generates masking attestation reports documenting the techniques applied, the columns masked, and the verification results. These reports are the evidence base for scope reduction claims in compliance assessments.

Automation and DevOps integration

Static masking should be part of your database refresh automation, not a manual step that gets forgotten. DBsafe provides a REST API for triggering masking jobs programmatically, enabling integration with your CI/CD pipelines, database provisioning scripts, and scheduled refresh workflows.

Integration examples: trigger a masking job as part of your nightly production database backup process, creating a fresh masked copy ready for development use each morning. Or trigger masking as part of a pull request environment provisioning workflow, ensuring every ephemeral environment contains masked data.

Platform Capabilities

Everything you need to address static data masking — in one integrated platform.

Full Database Masking

Mask an entire database schema in a single job, preserving structure and referential integrity.

Sensitive Column Discovery

Automatically identify columns requiring masking before the job runs.

Format-Preserving Substitution

Masked values match real-world formats: valid emails, Luhn-valid cards, correct postcodes.

Masking Attestation Report

Post-job verification confirms zero sensitive data remains, with a signed evidence report.

API-Driven Automation

Trigger masking jobs from CI/CD pipelines and scheduled refresh workflows via REST API.

Large-Scale Performance

Batch processing and parallel execution for databases containing hundreds of millions of rows.

Frequently Asked Questions

Answers to the most common questions about static data masking.

What is static data masking?

Static data masking is a technique for permanently replacing sensitive data in a copy of a database with realistic but non-sensitive substitutes. The original database is unchanged; a new masked version is created. Static masking is used for development, testing, and third-party data sharing scenarios.

Does static data masking affect the original database?

No. Static masking operates on a copy of the database. The source database is read in a non-destructive way, and the masked output is written to a separate target database. The production data is never modified by the masking process.

How long does a static masking job take?

Masking time depends on database size, number of sensitive columns, and server resources. DBsafe's parallel processing engine can typically mask a 100GB database in 30-60 minutes. Very large databases can be processed overnight as part of a regular refresh cycle.

Can static masking satisfy PCI DSS non-production data requirements?

PCI DSS Requirement 6.3.3 requires that production PANs be masked in non-production environments. DBsafe's static masking replaces all PANs with Luhn-valid dummy values. A masking attestation report documents the masking, supporting your QSA assessment. We recommend confirming scope reduction with your QSA.

What happens if new sensitive columns are added after the masking rules are configured?

DBsafe runs a discovery scan before each masking job and flags any new columns that may require masking. If a sensitive column is detected that is not covered by an existing masking rule, the job pauses and an alert is sent for review. This prevents unmasked sensitive data from appearing in the masked output.

Ready in 5 minutes

Refresh your development environments with safe, masked data

Static masking with DBsafe creates production-equivalent databases that your development team can use freely without compliance risk.