Quality Assurance & Test Automation in Digital Transformation

Importance of Testing:

The success of any company after digital transformation hinges heavily on its customers. During the transformation of any kind, organizations usually focus on upgrading their infrastructure and therefore remove existing obsolete processes and implement newer efficient ones instead. It is in this crucial period of transition that QA is indispensable.
Testing involves two words, Verification, and Validation. Whatever the product, be it regular toothpaste or software delivery, nothing goes into production without being verified and validated for use. For every software product (be it windows or web applications) the anomalies should be encountered early so that they can be rectified quickly, resulting in the delivery of polished, high-end products.

Testing teams must have the entire process covered from beginning to end including creation, development, marketing, distribution, and delivery.

Methods of Testing:

• Manual Testing – This conventional testing method involves no technical knowledge. Understanding the core business or motive of the application and traversing the application from the end-user perspective covering business scenarios
• Automation Testing – As programming languages started evolving, the testing methodologies started using these programming languages to make the test steps automated through coding. This automated testing method provided a faster way of verifying the process within the application.

Why Automation?

The practice of Manual Testing has been effective for a very long time in the Testing world. However, to sustain in today’s high-tech world that comes with pressures of reducing effort and cost, Manual testing has given up its timeworn effectiveness to the newbie Automation framework. Automation tools and their frameworks serve the technology needs of the modern business world.

However, it must be noted that Manual Testing is still effective for the scenarios involving exploratory testing, which is based on the testers’ experience or usability testing that tests the interactivity of the application. Manual testing is constructive for ad-hoc testing as well.

QA teams should ideally test automation repeatedly on different devices and operating systems to simplify and perfect their whole customer engagement ecosystem.

Automation frameworks are a one-time investment; once the framework is created it provides a high ROI to both the client and vendors. Today’s clients demand Automation tools to perform every type of testing. To meet this demand modern Automation tools are designed to not only support functional testing but also other testing methodologies listed below which increase productivity with much less effort.

• Regression Testing
• Load Testing
• Performance Testing

Seen here are some comparison points between Manual and Automation testing and how Automation testing overcomes the drawbacks of Manual testing and will prevail for a long time to come.

Automation Testing v/s Manual Testing

Test Effectiveness and Reliability

Manual Testing: Manual Testing is not accurate all the time. It can lead to manual errors; its reliability is a cause of concern.

Automation Testing: Automation testing is more reliable as it is performed through tools (UFT, Selenium, TOSCA, etc.) and scripts where no human interference is required.

Resourcing and Cost

Manual Testing: Manual Testing will require high human effort to test the flows resulting in an increase in costs to the company.

Automation Testing: This is done through software tools, where only the cost of the license (if any) is to be borne and which requires only one-time coding to run the test steps. These steps can be reused in different test scripts, making them reusable test cases. The ROI on any Automation tool is high.

Implementation of Automation Testing:

In the software testing field of today, Automation testing has the upper hand. That being so, the implementation methods of Automation testing have been diversified as presented below.

Listed below are details of all the Frameworks along with their pros and cons

Modular Testing Framework:

As the name suggests, larger test scripts are broken down into independent tests. The changes made in one module don’t affect the other module.

Pros:

• The framework introduces a high level of modularization, which leads to easy and cost-efficient maintenance.
• The framework is pretty much scalable.

If the changes are implemented in one part of the application, the test script representing only that part of the application needs to be fixed, leaving all the other parts untouched.

Cons:

• While implementing test scripts for each module separately, we embed the test data (Data with which we are supposed to perform testing) into the test scripts. Thus, whenever we are supposed to test with a different set of test data, the test scripts need to be manipulated.

Data-Driven Framework:

In this framework, the test data drives the script. Multiple sets of test data are identified covering multiple scenarios. The test data detects which scenario and its test cases will be running by the script. Test data is maintained in an excel sheet.

Pros:

• The most important feature of this framework is that it considerably reduces the total number of scripts required to cover all the possible combinations of test scenarios. Thus, a lesser amount of code is required to test a complete set of scenarios.
• Any change in the test data matrix would not hamper the test script code.
• Increases flexibility and maintainability
• A single test scenario can be executed by altering the test data values.

Cons:

• The process is complex and requires extra effort to provide the test data sources and reading mechanisms.
• Requires proficiency in a programming language that is being used to develop test scripts.

Keyword Driven Framework:

This framework is driven by the keywords provided during the time of execution. These keywords determine which scenario to be tested along with which test data to use for the execution.

Pros:

• In addition to the advantages provided by Data-Driven testing, the Keyword driven framework doesn’t require the user to have scripting knowledge, unlike Data-Driven Testing.
• A single keyword can be used across multiple test scripts.

Cons:

• The user is required to be well-versed with the Keyword creation mechanism to be able to efficiently leverage the benefits provided by the framework.
• The framework becomes complicated gradually as it grows and a number of new keywords are introduced.

Hybrid Testing Framework:

As the diagram suggests, this framework is a combination of all the above-mentioned frameworks. The best features of all frameworks are leveraged here. Thus, making it far more scalable and reusable as scripts are broken into individual test cases.

In the excel sheet, all the keywords and test data will be mentioned and the script will look at the keywords and identify the test data. Through this multiple test data can be tested using a single keyword.

Behaviour Driven Development Framework:

BDD framework allows test scripts to be designed in simple Gherkin language which becomes understandable by BAs and the other stakeholders. For this to be implemented, Cucumber or JDD can be leveraged along with Selenium.

Deciding on the correct automation process to implement

This becomes a quite show-down while judging a best automation framework to implement. If we go by the cost, then a simple keyword or data-driven framework will suffice the need. When we try implement something which can cost us less and the maintenance of script very much easier, then Behaviour Driven Framework (BDD Framework) along with Hybrid framework will be a great choice. BDD will make the scripts very much understandable by every stakeholder.

BDD framework when combined with the Hybrid framework increases the reusability of scripts. So, every time there is no need to write a redundant step while testing an E2E flow. We just need to call the respective functions from a different script. Reporting can be customized as per customer need using different tools like extent report and logging of errors can be done using Log4j.

These features enable the maintenance effort to decrease significantly, as automation testers can dig up the code very easily through logs and reports and debug them without a hassle.

Thus, its Automation Time.

By Rebonto Dey

Senior Pega Tester