Non-Regression testing, what is it?

March 20 09:00 2011 by Nataliia Vasylyna Print This Article

The article was updated in May 2018.

Making code changes, a developer may affect the proper work of the software. Even a small tweak can cause unexpected consequences or provoke new bugs to appear. In order to detect newly appeared problems, testers perform regression testing.

Regression testing is re-execution of the tests to make sure that code modifications don’t affect existing functionality. But sometimes it happens that there is not enough time or resources to execute regression tests. In this case, a testing team can check only those system modules where the modifications have been made. They omit a full regression testing. This is non-regression testing.

Non-regression testing is a technique aimed to verify whether a new or modified functionality operates correctly with the assumption that the previous functionality wasn’t affected. When applying non-regression testing, testers check only the evolving unit or module instead of the whole product, thus, save resources and time.

Steps of non-regression testing

Define software benchmark
Detect routines to simulate software functionality
Execute routines on benchmark and new release version of software
Process the obtained results

Non-regression testing can be included into regression checking. Here is an explanation how it works. Let’s imagine that there is a functionality A which have been already tested and a new functionality B just added to the product. So, for now, the program functionality is A+B. Non-regression testing will cover only B functionality. But later new functionality – C – will be implemented. Now if we want to perform regression testing, it will consist of the tests for A and B functionality. In this case, non-regression tests become a part of regression testing.