Anonymise, Pseudonymise or Mask Test Data

We offer automated procedures for processing test data regarding anonymisation. They can be used retrospectively, as so-called in-place processes, which allow the processing of mass data, and can also be directly integrated into the provisioning process of the test case data.

The PII Finder (personal info identifier) supports the process modeler in recognizing relevant columns. The software offers standardized methods for anonymization/pseudonymization/masking: first and last names, addresses, bank details, social security numbers, etc. The user can modify the methods or introduce their own. The same applies to the value ranges to be used, national / international address lists, etc.

Methods and rules are integrated in application models for company areas or divisions in such a way that the defined anonymization is carried out automatically, depending on the test bed to be supplied.

With our processes you anonymize or pseudonymize uniformly company-wide, whether Oracle, Db2, IMS, VSAM, LUW, MS SQL Server, PostgreSQL, SAP, cross-application.

Anonymization plays an important role in every test data management concept. Most of the time, production data must not be used for test purposes. Whether it be protection against competitors, outsourcing of test departments or compliance with the data protection law, all of these reasons require anonymization of the test data.

A professional concept for anonymizing test data is usually divided into various phases. Important preparation work must be done before the actual anonymization can take place.

Identification

The data model must be examined for sensitive columns, or attributes containing names, addresses, bank details, etc. must be identified. Without the support of efficient software, this phase can already be one of the biggest hurdles when implementing an anonymization strategy.

Implementation

Based on the columns that are identified, rules must be stored that anonymize the data according to the company specifications. Procedures for redistributing the previous data ensure that marginal cases that are contained in the productive data and are therefore valuable for a test continue to exist. The consistent anonymization across tables or system boundaries must also be taken into account.