Testing Grid

WHO Who is the stakeholder/interested party? Developer Current
Develpoer Future
Developer Current
Tester Current
Developer Current
Tester Current
End User/Consumer/End Point
(Also known as the 3 Amigos)
WHAT What is tested? Components/Services/Units Plumbing/Connectivity Does what the stakeholder expects
What level of detail? As much as possible to narrow down errors/defects Works end to end without issues Works end to end, data is as expected
What Code Coverage? 100% (Or as close to as possible – within reason)
– Error Paths
-Edge Cases
50% Enough to prove working 20-30% Enough to see that the requirements have been met.
WHEN When executed? – Development
– Regression
– Bug Fixing
– Working out what the code does
When more than 2 components are complete
(Currently done after development)
– Once End-To-End complete (Even partial)
(Currently done after SIT)
WHY Why is it done? – Verification Of Unit
– Boundary Conditions
– Regression
– Verify End-To-End
– Regression
-Verify works as customer expects
WHERE Where is it tested? – Development Environment – Build Environment
– Test Environment
– Stage Environment
– Preprod Environment
HOW How is it tested? -NUnit (Junit/xmlUnit/WMUnit etc)
– Stubs
– Mocks
– Virtual Services (CA DevTest, IBM Rational Tester etc)
– Commercial Frameworks
Test Driven Development(TDD)
-NUnit (JUnit, xmlUnit, WMUnit etc)
– Virtual Services
– Mocks
– Commercial Frameworks
– Actial End Points
– NUnit (Junit, xmlUnit, WMUnit etc)
– Commercial Frameworks
– Actual End Points
– Behaviour Driven Development (BDD), Acceptance Test Driven Development (ATDD)
How long should a test take to run? Seconds Seconds -> minutes minutes -> hours

Thoughts, Tutorials and Learnings specifically with regards to Agile, DevOps and SOA/EAI