Part 3: How to Develop a Data Integration Master Test Plan
In the final data integration series article, we'll show you how to develop a data integration master test plan, the cornerstone of data verification efforts.
Join the DZone community and get the full member experience.Join For Free
In part one of this 3-part series, we covered why assessing risks early and often is key, the best practices for addressing risks, and best practices for common risk mitigation. Part two covered examples of quality risks for integration projects and best practices for tackling them. In the final article of this series, we'll show you how to develop a strong data integration master test plan.
Although Agile testing tends to deprioritize test planning, teams working on data integration projects would be remiss to overlook the long-standing motives and rationale for a project-wide, data integration master test plan (MTP).
A “Data Integration Master Test Plan” (MTP) represents a plan of action and processes designed to accomplish quality assurance from the beginning to the end of a data integration development lifecycle. The test plan should describe all planned quality assurance for each SDLC phase and how QA will be managed across all levels of testing (ex., unit, component, integration, system testing, etc.). The MTP provides a project-wide, high-level view of the quality assurance policies (often based on IEEE Standard 829).
Such a plan may be developed using the data project documentation
- Business and technical requirements
- Data dictionaries and catalogs
- Data models for source and target schemas
- Data mappings
- ETL and BI/analytics application specifications
It’s essential to purge the data integration target data of the most severe and disruptive bugs. The sooner data quality/testing objectives are defined, the better your chances of exposing issues early when they’re easier, faster, and less costly to fix.
Why Develop a Master Test Plan?
Give your developers a standard test plan document that lays out a logical sequence of actions to take when performing integration tests. Doing so keeps testing consistent across the project and allows project managers to allocate the right resources to begin the integration testing process.
A data integration MTP should describe the testing strategy/approach for the entire data integration and project lifecycle. The MTP will help the project team plan and carry out all test activities, evaluate the quality of test activities, and manage those test activities to successful completion.
The MTP should be published and distributed for approval by business and technical stakeholders to inform everyone about crucial areas of the planned data integration testing process:
- Testing and quality objectives
- Scope and constraints of planned testing
- Test environments
- Test data sources
- Testing methods
- QA tools, processes, and schedules
- Testing resources required for the project
The MTP should also summarize the test team’s objectives for:
- Work products
- Testing procedures
- Testing assumptions
- Project risks
- QA entry and exit criteria
- Testing roles and responsibilities (including those of business analysts, developers, users, etc.)
- Defect tracking and reporting processes
- Change control processes
Types of Data Integration Tests
The types of quality assurance verifications conducted during unit, integration, system, and acceptance tests (described in the data integration MTP) should be the following:
- Business requirement verifications
- ETL testing
- Data integration testing (record counts, data quality, performance, etc.)
- System integration testing
- Functional and non-functional business requirements
- Technical requirements testing
Another focus of the data integration MTP is an end-to-end data integration test process, including validating:
- The loading of all required data integration databases and tables
- The correct execution of all data transformation and cleansing according to business rules and reporting requirements
- Successful completion of data aggregations
Reduce Risks While Planning for Tests
The MTP should describe known testing risks/challenges and plan a testing approach to address each. Doing so will serve as an essential aid in test planning. Common risks include:
- Lack of detail in requirements documents
- Frequently-changing business and technical requirements
- Heterogeneous, complex, and massive data movement
- Missing and duplicate data to be identified, corrected, and tested
- Data that must be transformed and cleansed (which often results in sophisticated testing)
- Lack of access to commercial or open-source data integration testing tools
An MTP should document major priorities for reducing business and technical risks associated with the development, deployment, and operation of the data integration. Using appropriate test cases and allocating sufficient testing resources will significantly reduce risks and make testing more effective and efficient. The MTP will enable the project teams to develop a superior product and lay a strong foundation for each iteration.
A Summary of Skills Needed for Data Integration Testing
The MTP should demonstrate that data integration testing is a unique endeavor that requires specific knowledge and skills:
- A firm understanding of data structures, analytics, and database concepts
- Advanced expertise with database queries
- Expert data profiling methods and tool skills
- Experience with MS Excel data analysis functions
- Skills to develop data integration test plans
- Anticipating Data Integration Testing Challenges
An effective Data Integration Master Test Plan is the cornerstone of the entire data integration verification effort. The MTP guides test engineers as they develop tests, ETL’s, integration, performance, security, end-to-end, and application testing. A good quality MTP will help stakeholders understand what will be tested (and how), as well as offer assurance that hight quality was carefully considered.
A master test plan that was successfully implemented on a data integration (data warehouse) project can be found here: https://www.academia.edu/39634008/Project_Master_Test_Plan_MTP_V
Opinions expressed by DZone contributors are their own.
Beginner Intro to Real-Time Debugging for Mobile Apps: Tools and Techniques
Building Resilient Systems With Chaos Engineering
Automated Multi-Repo IBM App Connect Enterprise BAR Builds
What Is End-To-End Testing? E2E Testing Tutorial With Examples and Best Practices