Creating Testing work practice
The testing work practice outlines the testing phase of the project. The purpose of this work practice is to highlight the major deliverables/products and processes. The testing phase defines how deliverables will be assessed, and defects corrected. This work practice does not cater to all the different testing that can be performed on a project, but rather for testing specific components of the project.
Getting ready
Before starting, it is important to have an understanding of your project methodology and determine:
- The existing work practices
- The key roles which will be involved in the work practice
- If testing is executed within your projects
How to do it...
Work practices are best created as a visual tool. To do this, start up a diagramming tool such as Microsoft Visio or Microsoft PowerPoint.
- In the diagramming application, create a new tab or slide called Testing.
- On the diagram, create swimlanes. In each swimlane, add the key role.
- Step through this phase of the project logically, and add the key products/deliverables and processes.
- A sample testing work practice is shown here:
- Once you have created your products and processes, you will need to describe each of them.
For example, System/Functional Test Cases are test scripts outlining the ways to test a component.
- Once all the products and processes are described, you can determine the deliverables for each product and the role responsible for delivering it, based on the swimlane.
For example, the deliverable for System/Functional Test Cases is a set of instructions on how to test a component, and the expected outcome created by Testers.
How it works...
Work practices give members of the project a quick and easy way to understand the flow of activities within the phase, and all the deliverables required with the phase.
There's more...
The testing phase may not be a defined phase in your project but may be spread among the other phases. This phase is generally broken down to account for effort and to aid in estimation. Additional testing on the project will be required, such as:
- Integration Testing — also known as black box testing, focuses on the outcomes of the process
- Performance Testing — focuses on the capabilities and thresholds of a solution
- Unit Testing — also known as white box testing, focuses on tracing the inputs and outputs as they are processed by a unit, to validate whether the information outcomes are correct
As the data in business intelligence and data warehouse solutions can utilize large data sets, it is important to spend some time to prepare the positive and negative data sets.
Positive data sets are based on production quality data, which can be placed through the solution to determine outcomes. This may not be actual production data as it may be stripped off of the sensitive information.
Negative data sets are based on production quality data, the same as the positive data set. The difference is that this data set is intentionally altered to included errors and anomalies, to force errors and warnings within the solution.
It is important to understand that due to the vast quantities of data in the solution, it may not be feasible to test all scenarios.