What is regression testing?

It's the practice of re-running existing tests after a code change to make sure previously working functionality hasn't been broken.

What is a 'regression' in software?

A regression is a bug where something that used to work correctly stops working, usually as an unintended side effect of a code change.

Is regression testing manual or automated?

It's almost always automated in modern development, since manually re-verifying the whole application after every change doesn't scale.

How does a regression suite grow over time?

Teams commonly add a new test for every bug that's fixed, specifically reproducing the original issue so it can be caught immediately if it ever reappears.

How is regression testing related to the test pyramid?

Regression coverage typically follows the test pyramid's shape, with the bulk of tests being fast unit tests, supplemented by fewer integration and end-to-end tests.

Why does regression test run time matter?

As a suite grows, slow regression tests can bottleneck CI/CD pipelines, so teams manage run time through parallelization, selective test execution, and pruning low-value tests.

DevOps

Regression Testing

BeginnerTechnique6.4K learners

Regression testing is the practice of re-running existing tests after a code change to confirm that previously working functionality has not been broken by the change.

#RegressionTesting #DevOps#Technique#Beginner#TestPyramid #TestCoverage #MutationTesting #SmokeTesting #SoftwareDelivery#Glossary#SkillVeris

Definition

Regression testing is the practice of re-running existing tests after a code change to confirm that previously working functionality has not been broken by the change.

Overview

Every time code is modified — whether to add a feature, fix a bug, or refactor internals — there's a risk that the change unintentionally breaks something that used to work, a phenomenon commonly called a 'regression.' Regression testing exists specifically to catch that risk: by maintaining a suite of tests covering previously verified behavior and re-running it after every change, a team gets confidence that new work hasn't silently broken old functionality. In modern software development, regression tests are almost always automated rather than manually re-executed, precisely because manual re-testing of the entire application after every change doesn't scale. A regression suite typically accumulates over time: whenever a bug is found and fixed, a new test is added that specifically reproduces the original bug, ensuring it can never silently reappear (sometimes called a 'bug regression test'). The suite also includes broader tests covering core functionality across unit, integration, and end-to-end layers, following the shape of the test pyramid so that the majority of regression coverage comes from fast, cheap unit tests rather than slow end-to-end tests. Regression testing is what makes continuous integration trustworthy — without a comprehensive, fast-running regression suite, teams either slow down releases to manually verify nothing broke, or ship changes blind and discover regressions from users in production. As a codebase and its test suite both grow, managing regression test run time becomes its own engineering problem, commonly addressed through parallelization, selectively running only tests affected by a given change, and periodically pruning redundant or low-value tests.

Key Concepts

Re-runs previously passing tests after a code change to catch unintended breakage
Almost always automated in modern development workflows rather than manual
Grows over time as new tests are added for each bug fix, preventing recurrence
Spans unit, integration, and end-to-end tests following the test pyramid
Central to making continuous integration and continuous deployment trustworthy
Requires ongoing management of test suite run time as the codebase grows
Distinct from smoke testing, which is a narrower, faster sanity check

Use Cases

Confirming existing functionality still works after adding a new feature

Preventing previously fixed bugs from silently reappearing

Validating refactors haven't changed observable application behavior

Gating merges and deployments in CI/CD pipelines

Building confidence for frequent, incremental releases

Supporting safe large-scale codebase migrations or dependency upgrades

Frequently Asked Questions

From the Blog

Programming

Testing Python Code with pytest: A Beginner's Guide

Untested code is legacy code from the moment it's written. This guide explains how to write effective Python tests with pytest — from your first test function through fixtures, parametrize, mocking, and measuring coverage.

Scikit-Learn for Beginners: Machine Learning in Python

Scikit-learn is the most widely used Python library for classical machine learning. This guide covers the fit-predict workflow, train/test splits, classification, regression, model evaluation, feature engineering, and pipelines — everything you need to build and evaluate your first ML models.