Software Testing Social Network

Acceptance Testing is a distinct pre-delivery stage that involves transfer of application ownership by the developers to actual end-users, hence called UAT (User Acceptance Testing) or CAT (Customer Acceptance Testing).

Objective of Acceptance Testing

• It is driven by the objective that developed software must satisfy the stated requirements and meet customer/user expectations. The application must be fit for business purpose, which is called validation.
• The objective of acceptance testing is not to look for faults actively.

Requirements/considerations  for Acceptance Testing 

• Before entering this stage, it is expected that in the preceding stages software faults have been identified and therewith resolved. It is considered that system is virtually ready for operational use.
• This testing phase must not seek to replicate other testing phases.

During this phase it is important to reduce system changes, as the rework cost is high in this stage.

• Ideally, during acceptance testing the software under test, must be taken as ‘frozen’ as implementing changes at this stage can invalidate testing which have already been conducted.
• People involved in acceptance testing must dictate objectives and consider the following –
• Apt coverage of each business function.
• Appropriate coverage of system reports and screens.
• Demonstration of acceptance test conditions.
• Realistic business scenarios for test conditions model.
• Apt coverage of application menu options.
• Up-dation of system documentation and training materials.
• Acceptance testing needs higher levels of regression testing.
• If system flaws are not exposed during acceptance testing ‘work-arounds’ must be investigated before undertaking any change to the software.
• But correction with high severity/priority must be made. 
• The steps involving system investigation, system review, agreement and/or deferral are repeated until a point is reached where system is considered as acceptable.  

Significance of Acceptance Testing

• Acceptance Testing gives an opportunity for reviewing system effectiveness and integrity. Any loopholes or problems found in the system leave scope for improvement.
• At times existing test scenarios are ran for providing confidence that previous test phases have been appropriately completed.
• Things to Remember before Acceptance Testing
• Acceptance testing must not start at point of software delivery to minimize the risk of software faults and system changes. For this users must be involved in the system testing at an early design stages.
• Timely involvement of users in testing ensures avoidance of re-work and reduced cost of system delivery.

Comments (0)

Subscribe to this comment's feed