DATAVAULT — TEST DATA MANAGEMENT

Enterprise Test Data Management for Connected Applications

Discover valid enterprise data, understand its business relationships, protect sensitive values and supply reusable test datasets to implementation, regression and autonomous testing.

Request Test Data Demo Explore DataVault

Automated testing is only reliable when the right business data exists in the target environment.

DataVault — Test Data Lifecycle
Discover
↓
Organize
↓
Protect
↓
Select
↓
Reuse
↓
Supply

Test Automation Often Fails Because the Data Is Wrong

Automated testing is only reliable when the right business data exists in the target environment. A test script can be correctly built and still fail — not because the application changed, but because the record it depends on no longer qualifies for the scenario.

Supplier not active
Customer account missing
Worker not eligible
Purchase order not open
Accounting period closed
Bank account invalid
Currency not enabled
Payment term unavailable
Prerequisite object missing
Related records inconsistent

Static, hard-coded test data becomes stale as configuration, master data and business rules change over time — a supplier that was valid last quarter, a period that has since closed, a payment term that has been retired. DataVault is designed to keep test data grounded in what is currently valid in the connected system.

Understand Business Objects and Their Dependencies

Enterprise applications are not flat tables — a supplier depends on a supplier site, a worker depends on an organization and assignment context, an opportunity depends on an account and contact. DataVault connectors are designed to understand these application-specific relationships rather than treating every system as an unrelated list of records.

Oracle

Supplier
↓
Supplier Site
↓
Bank Account
↓
Payment Method

Workday

Conceptual
Worker
↓
Organization
↓
Assignment
↓
Payroll / Compensation Context

Salesforce

Account
↓
Contact
↓
Opportunity

SAP

Conceptual
Business Partner
↓
Company / Purchasing Context
↓
Financial Relationships
Screenshot of the Syntra DataVault object browser listing business modules such as Absence Management, Accounts Payable and Core HR, with the Locations object expanded showing 1,750 harvested records and options to export to Excel or JSON.
Example object browser from a connected Oracle Fusion test environment — DataVault organizes harvested records by business module and object.

Turn Enterprise Data Into Test Dimensions

DataVault is designed to organize harvested data along the dimensions testers actually plan scenarios around, so a single business object can be filtered and combined the way a test case needs it.

Supplier Business Unit Currency Payment Terms Account Tax Invoice Type Amount Date
DataVault
Test dimensions
→
Jarvis
Scenario generation
→
Scenario Combinations

This directly relates to AI-generated test coverage — Jarvis can draw on DataVault dimensions to build scenario variations instead of relying on a single fixed data combination.

Select Data Based on the Test Requirement

Testers typically describe the data they need in terms of the business condition it must satisfy, not a record ID. DataVault's scenario and data-selection logic is designed to work from criteria like this:

"Find an active supplier with a valid supplier site, GBP currency and valid payment terms in the UK Business Unit."

This represents the underlying selection criteria model DataVault evaluates against harvested, organized data — where matching data exists, DataVault is designed to return it. It is not a natural-language chat interface; it is the logic that scenario and data-selection features are built on.

Build Reusable Data Packs

AP Invoice Dataset

  • Business Unit
  • Supplier
  • Supplier Site
  • Currency
  • Payment Terms
  • Distribution

AR Invoice Dataset

  • Business Unit
  • Customer
  • Transaction Type
  • Currency
  • Payment Terms
  • Revenue Account

Worker Dataset

  • Worker
  • Organization
  • Location
  • Manager / Reporting Context
  • Relevant Employment Attributes

Data packs are designed to be reused across implementation cycles, regression runs and autonomous test execution rather than rebuilt for every test pass.

Use Realistic Data Without Exposing Sensitive Values

Realistic test data and privacy are not a trade-off. DataVault is designed to combine test-data usability — valid, dependency-correct business records — with masking policies applied to the sensitive attributes those records carry.

Explore Data Masking →
Enterprise Data
↓
DataVault
↓
Select valid relationships
Mask sensitive attributes
↓
Privacy-Protected Test Dataset

Use the Data You Need — Not an Entire Production Copy

Concept

This is a concept DataVault's architecture is designed around, not a claim of a fully shipped, generally available feature today: select a representative business subset and retain the dependencies required for testing, rather than working with — or copying — an entire production dataset.

Worker Subset Example

500 workers, plus the organizations, managers, payroll context and related reference data required to keep those workers testable.

Supplier Subset Example

200 suppliers, plus their sites, bank and payment configuration, currencies and payment terms.

Detect When Test Data Becomes Stale

Data that was valid when it was harvested does not necessarily stay valid. DataVault is designed to be run to identify conditions such as:

Supplier disabled
Customer status changed
Accounting period closed
Worker terminated
Purchase order consumed / closed
Bank account changed
Reference data removed

Where configured, DataVault can be run to refresh and revalidate harvested data against current source-system conditions, helping flag datasets that need to be re-selected before they are relied on for a test run — this is a scheduled or on-demand check rather than an automatic, instant, real-time guarantee.

Turn Test Data Into AI-Generated Test Coverage

Jarvis can use DataVault dimensions to generate scenario variations from a standard business test, turning one manually authored test into a broader set of positive, negative, boundary and configuration-specific scenarios.

DATAVAULT — Valid + protected test data
↓
JARVIS — Positive / negative / boundary variations
↓
REGRESSION PACK
↓
SYNTRAFLOW — Scheduled batch execution

Standard Test

"Create Supplier Invoice"

DataVault Dimensions

Suppliers, Business Units, Currencies, Payment Terms, Accounts, Amounts

Jarvis Generates

  • Positive combinations
  • Negative conditions
  • Boundary scenarios
  • Configuration-specific variants

Reuse Data Across Scheduled Regression Packs

DataVault's relationship to SyntraFlow regression testing follows a consistent sequence:

  1. Select scenario
  2. Select data dimensions
  3. Generate variations
  4. Add to regression pack
  5. Run now
  6. Schedule
  7. Collect evidence

Manage Data for End-to-End Tests

Cross-system tests require corresponding business entities and consistent masked identities on both sides of the flow — the same worker or supplier needs to be recognizable, and correctly masked, at every step.

Workday Example

Worker
↓
Payroll
↓
Finance
↓
Data Lake

Oracle Example

Supplier
↓
Payment
↓
Bank
↓
Reconciliation

Tracing which downstream systems a business entity reaches, and confirming its masked identity is applied consistently along the way, is covered in more depth on Data Lineage and Downstream Masking.

DataVault Across Enterprise Applications

DataVault is built on a connector-based architecture. Platform-specific pages are published as connectors are validated.

Workday →

Worker and implementation data masking and downstream verification.

Oracle Fusion →

Object-level test data harvesting and masking for Oracle Fusion environments.

SAP

Roadmap

Connector not yet validated.

Salesforce

Roadmap

Connector not yet validated.

Frequently Asked Questions

What is enterprise test data management?
Enterprise test data management is the practice of discovering, organizing, protecting and supplying valid business data so implementation and automated tests can run against realistic, dependency-correct scenarios rather than incomplete or stale records.
How is DataVault different from a spreadsheet of test data?
A spreadsheet captures a static list of values at a point in time. DataVault is designed to understand business objects and their dependencies within a connected application, apply masking policies to sensitive attributes, and support scenario-based selection logic so testers work with data that reflects current, valid business relationships rather than a list that can go stale.
Can DataVault keep test data private?
Yes. DataVault is designed to combine test data selection with configurable masking policies, so sensitive attributes such as personal, financial or payroll details can be protected while the underlying business relationships needed for testing are preserved. See Data Masking for details on policy configuration.
Can test data be reused across environments?
Datasets and scenario/data selection logic are designed to be reusable across implementation, regression and autonomous testing where the same objects and connectors are configured, subject to authorized connectivity for each environment.
Can DataVault provide data to Jarvis?
Yes. Jarvis can use DataVault's test dimensions — such as supplier, business unit, currency, payment terms and account — to generate positive, negative, boundary and configuration-specific scenario variations from a standard business test.
Can DataVault manage cross-system test data?
DataVault is designed to help testers work with corresponding business entities and consistent masked identities across connected systems, supporting end-to-end scenarios such as Workday worker-to-finance flows or Oracle supplier-to-payment flows where the relevant connectors are configured. See Data Lineage and Downstream Masking for related capabilities.
Does DataVault copy production data?
DataVault is designed to consume, reference and select test data according to the configured architecture and applicable supported connectors, rather than perform unrestricted or automatic production data extraction. Data access is subject to authorized connectivity and configuration in each environment.

See DataVault Across Your Enterprise Landscape

Show us a representative application and downstream data flow. We'll demonstrate how DataVault can model sensitive data, masking policies, test-data relationships, lineage and downstream verification.

Request DataVault Demo Explore DataVault