At its simplest, testing means finding out if something is working or not. For example, if you purchase a light bulb, you fit it to test the filaments work, and the lamp ignites. Similarly, Software Testing involves testing software to ensure the application is working as expected or not. In this post, we take a look at the types of testing in software engineering.
Manual and Automation Testing
Although the term “Software Testing” sounds pretty simple, it can be quite complicated at times because of the array of techniques and types of testing available to the software engineer. Whichever tests you select will come from one of two categories, either manual or automated, where test scripts are used with a choice of tools. Manual testing is seen as easier to do because it doesn’t involve having to learn how to use automated testing tools, but the repetitive nature of the work can be very tedious. Alternately, though automation involves more costs leading up to implementation, it does offer benefits when it comes to executing repetitive tasks. Some projects, however, apply both manual and automated testing methods to achieve optimal test results.
Types of Software Testing
In software testing, testers refer to detailed business and functional requirement specifications, which then enables them to identify deviations between actual and expected behavior. These discovered deviations are reported as bugs to be fixed by developers, then once again verified by testers. During this whole process, testers run different types of tests. The scope of testing and types of tests will differ based on requirements and the kind of project. There are also different types of tests run at various stages of the project. Let’s take a look at the most prevalent types of testing conducted during the overall software development lifecycle.
During the Development & QA Phase
This list outlines some tests run during the development and QA phase, irrespective of the methodology followed for software development such as Agile or Waterfall.
- Unit/Component Testing is mostly done by developers to check the unit-level code where a unit represents the smallest piece of testable code. The Unit Test is the first level check done before any other testing.
- Sanity and Smoke Testing are build related. When a new code is ready for QA, the testers run a quick sanity check to confirm and accept the build/release for further testing.
- Functional Testing is the primary requirement of all projects. In functional testing, the tester examines every functionality in detail. A tester refers to test cases for running the functional tests.
- Regression Testing is a type of test which is necessary for any software testing project. Regression testing is performed post-bug fix. This type of analysis is undertaken to ensure other functionalities have remained intact following the bug fix. One project might require several rounds of regression testing, depending on the bug release process.
- Integration Testing is done when the application interacts with several other systems. Testing is done to ensure integrations between systems function correctly, and data flows in and out without any errors.
- System Testing is a check to ensure the complete end-to-end system works as expected.
- Browser Compatibility is a check done by running the same application on different browsers/devices to ensure that the program/application behaves in the same manner, irrespective of the browser/device used by end-users.
- Localization Testing is required when the program/application build is expected to be run in several regions. Regions differ based on currency, timezone, language, and other factors. Therefore, as part of localization testing, the tester primarily focuses and runs tests on these parameters.
- Security Testing is done to identify any vulnerability or weakness in the system which can expose confidential data and resources to intruders. Software applications can contain a lot of personal user data in the database, and it is crucial to abide by the current security policies and be security compliant. Security testing uses tools.
- Performance Testing. In terms of speed and responsiveness, an application’s performance is reviewed using performance testing tools before updates are made based on recommendations. However, for website testing, the performance check is mainly run once the site goes live so an accurate assessment can be made. Testing in staging or development usually doesn’t reveal the correct performance for page load time due to there being a shared server.
- Load Testing is a type of non-functional test conducted to find out how the application performs under a specific load. Load testing is required for applications where the expected number of end-users is beyond what is normal.
- Exploratory Testing is the most preferred type of testing by testers. Testing takes place without test cases being written, and the tester has the choice of testing on the fly with the core objective being to touch each area and try to break the system.
During UAT Phase
User Acceptance Testing (UAT) occurs after development but before going live.
- User Acceptance Testing occurs after QA is complete and approved. UAT is primarily done by the business users to check if the built application satisfies the user requirements. At times, testers support and assist the stakeholders in performing UAT.
- Beta Testing is a type of UAT in which a limited set of end-users tests the developed application. The purpose is to observe how the application behaves in a contrived real-life setting before releasing the system for use by everyone.
After Going Live
- Post Production Testing – Although we often hear that all testing must be done during production, the reality is that post-launch, a set of tests need to be run to ensure that deployment went smoothly and did not break anything. Post-production is not the time for rigorous testing, but a high-level assessment using a checklist must take place.
- Maintenance Testing occurs after an application is live. Any testing of enhancements or fixes that take place after a product launch is termed as maintenance testing.
Conclusion
Determining which type of testing to run for your needs can be challenging, but once you understand the different kinds and how they fit into the overall scheme, making your choice easier. Keep in mind that you need to know your needs thoroughly before you make a decision.
