As the complexity of modern solutions increases, not only does the challenge for software developers grow, but so do the expectations for software testers. Although values such as ‘lines of code’ are not universally meaningful metrics, the following graph shows well where the journey is heading.

In the early days of computer technology, programs were relatively simple and straightforward. However, as technology advanced and software applications became more powerful, the size of code bases increased significantly. This is nothing wrong, of course, but it does provide new challenges.

Increased complexity – more bugs

As software has evolved, so has its complexity. Software systems now encompass complex architectures, sophisticated algorithms, and diverse integrations with various platforms and technologies. The demand for advanced functionalities, seamless user experiences, and robust security measures has led to the development of complex software applications. Managing this complexity requires careful design, modularization, and the adoption of architectural patterns to ensure maintainability and scalability.

While software has become more sophisticated and powerful, the presence of bugs remains a challenge. As software grows in size and complexity, the potential for defects and errors also increases. The sheer volume of code, coupled with interdependencies between modules and integration with external components, introduces a higher likelihood of bugs.

The need for faster product delivery can impact product quality

Companies are increasingly under fire to deliver software with new functionalities and improvements to stay ahead of the competition. This puts pressure on all involved roles to deploy new code rapidly, which often results in reduced testing efforts and even more bugs in production.

In today’s competitive business environment, a company’s reputation is also closely tied to the perceived and experienced reliability of its services. Even the smallest of bugs can have a significant negative impact on a company. In conclusion, testing is a crucial discipline to identify bugs before they are discovered by the end-users to secure the company’s reputation.

The crucial role of testers in product quality assurance

In this game, testers play a crucial role in safeguarding product quality by ensuring that the product meets the required standards in terms of functionality, reliability, usability, and more. This can’t be achieved just by clicking around in an application and reporting some “strange behaviors”. Testing should be a comprehensible activity based on specification, proven methods, and an overall understanding of the business context.

To ensure product functionality and quality, you need skilled people with a quality-focused mindset, who understand business processes as well as the implemented technologies and solutions. This work cannot be a “last minute” task that can be skipped in case of tight deadlines or done by any person “just available”.

Shift testing to the left and involve testers from the beginning

So, how is it possible to implement quality assurance in such a way that it does not delay the delivery date and still has an actual effect?

This requires involving experienced testers who have the capability to analyze the use cases and IT solution design. Even more crucially, they should be involved from the very beginning of the process. By challenging business ideas and identifying missed elements or flows early on and even without code, testers can help avoid wasting unnecessary resources on functionalities that may need to be changed later. This calls for close collaboration between business, testers, and developers as one team.

Starting testing early in the software development life cycle (SDLC) is known as the “Shift left” approach, and it comes from the idea of shifting test activities to the “left” in the development process timeline to ensure problems are detected as early as possible. Involving testers right away, starting with the requirements gathering stage, brings numerous benefits, such as early defect detection, faster feedback to developers, better collaboration between teams, and reduced business risk.

Efficient testing without excessive costs

In the realm of rapid code deployment, financial investments alone won’t solve quality issues. In addition to “Shift left” there are different methods that can ensure successful testing efforts without increasing the testing budget.

Testing isn’t just about speed; it’s a strategic blend of prioritization and disciplined collaboration. At Sixsentix, we advocate a disciplined testing approach centered on business-facing testing and business risk coverage.

With our risk-based methodology, we can clearly identify the processes and features with the greatest business impact, based on the frequency of use and potential damage that can occur as a result of failure. We perform a thorough analysis of the business, accurately determine specific business risks thanks to the combination of expertise and domain knowledge, and then prioritize testing of sensitive components.

This way, we make sure that key functionalities are swiftly checked and validated, and that the most pressing business risks are ruled out. Sixsentix combines this approach with the “Shift left”, which allows continuous testing and deployment after every change. This means that the latest updates are always released as soon as possible, without critical failures that could damage your reputation and business long-term.

Enhancing software quality through team dynamics

It’s also crucial to have a cohesive team composition, backed by a time-tested method. Our testing teams have been using our well-refined SWAT methodology, which enables them to use a systemic approach to testing.

Simultaneously, we abide by the “Tres Amigos” principle. It highlights the participation of three key roles: the product owner, the developer, and the tester, and emphasizes that these three roles are, in fact, one team, pursuing the same goals and priorities, which significantly boost the delivery of high-quality software.

Finally, in order to orchestrate and automate testing activities efficiently, we also designate a test architect for each project to define the overall testing strategy and framework and ensure they resonate with business objectives. This orchestrated effort significantly enhances the testing process, elevating software quality and ensuring a successful, reliable product.

Moving forward

Transforming quality assurance involves continuous improvement, robust collaboration, and the embrace of automation, propelling organizations towards a mature QA level where testing adapts to evolving requirements and aligns with changing organizational needs.

Let’s embrace efficient testing strategies and disciplined collaboration, striving for a future where software isn’t just fast and functional but also robust, reliable, and aligned with both user needs and business goals.

Author

Sixsentix

Sixsentix is a leading provider of Software Testing Services, QA Visual Analytics and Reporting, helping enterprises to accelerate their Software Delivery. Our unique risk-based Testing and QACube ALM Reporting and Dashboards, provide business with unprecedented quality and transparency across Software Delivery projects for faster time-to-market. Sixsentix customers include the largest banks, financial services, insurance, telecom providers and others. Sixsentix Onsite, and Nearshore (SWAT) services deliver optimized testing outcomes at significantly lower costs and help customers with scalability to keep pace with digitalization.

Sixsentix is an EXPO Exhibitor at EuroSTAR 2024, join us in Stockholm.

Let’s be honest. The dream of 100 percent test automation often turns out to be a nightmare. After all, it’s not just about writing a few scripts. Successful test automation needs to be well thought out, requires a test architecture and, above all, time. If you let the reins slip for even a few iterations, technical debt will creep in. In the worst case, the test cases become flaky. 

Less spectacular, but just as painful in the long run, is gentle erosion. While the system to be tested is constantly evolving, the automated scripts lag behind. They may still run successfully, but over time they become less and less meaningful.

“Not true!”, you are probably thinking. “We do test-driven development.” That’s laudable, but probably only part of the truth. Because you certainly have them too – the functional manual testers who cover the upper part of the test pyramid and look at the system as a whole, perhaps even in interaction with other applications. 

The Importance of Functional Testing

Functional testing is important. Of course, as much as possible should be checked during unit testing, but to be honest: the nasty bugs are hidden between the components. An example:

An app requires 2-factor authentication for login. Therefore, an 8-digit code consisting of letters and numbers is sent by email. Unfortunately, the app only allows numbers to be entered.

This example is not made up and illustrates the difference between unit and system testing. Each component complied with its specification and was tested successfully. Nevertheless, the overall system is unusable because the error was not in the code, but in the specification. Functional testers find such errors and that is what makes them so valuable.

Manual testing has never been easy, but today, in the age of Agile, it’s a pure race against time. If they don’t want to be left behind, they need to understand the intricacies of automated functional testing. Because test automation relieves them of the repetitive regression tests and allows them to focus on the important things, such as new features.

Fundamental Principles of Effective Functional Test Design

In fact, there are several common principles that are important in both manual testing and test automation:

• Clarity : Clarity is key. Structured, easy-to-grasp tests improve comprehension and minimize ambiguity, benefiting all stakeholders. Visual diagrams have long helped developers simplify complex problems. Testing can benefit from similar visualization, as functional tests mirror the intricate nature of systems under test. Visual test design definitely improves clarity, making it easier for test automation engineers to understand the business aspects of functional testing.

• Modularity : By breaking down complex scenarios into manageable test cases, manual testers can lay the groundwork for seamless automation, ensuring that each test remains a valuable asset throughout the software development lifecycle.
  
• Maintainability : Functional tests will continue to evolve, as will the associated test scripts. Some changes affect the technical level, others the functional level. Keyword-driven testing is a proven method of separating these two levels. Manual testers can thus contribute to the maintenance of automated tests without having to program themselves.

The goal of an effective, functional test design must therefore be to develop tests that are easy to maintain and update.

Transitioning from Manual to Automated Execution

Automation is often viewed as process optimization, as it takes over error-prone, repetitive tasks. Testers can use the time gained to focus on those tasks that require human judgment. Executing existing tests repeatedly following the accelerating rhythm of testing cycles is certainly not a mentally demanding task. The dumb writing down of test procedures is not a great intellectual accomplishment either. The real value lies in the preliminary considerations: What situations could occur? How should the system behave in this case? What can I do to push the system beyond its limits? Automation is therefore also a helpful and welcome support when creating test cases.

Studies[1] show that the combination of intellectual performance in test design and automated test execution not only increases test speed, but also the coverage of requirements and code. In general, the reliability of tests is increased when automated scripts support manual testers, especially in lengthy tests. This can be seen very clearly in load and performance tests, which are unthinkable without automation.

An intelligent automation strategy effectively balances human expertise with automated tasks.

A Practical Case Study – Yest Augmented by Maveryx

In this example, we show the successful marriage of two concepts represented by Yest and Maveryx.

Yest is a visual test design tool that implements a modern form of model-based testing (MBT) and test generation. Yest offers a whole range of functions to enhance the creative work of test design and speed up the painstaking task of writing test cases. Yest itself is agnostic, meaning that the generated scenarios may be used both for manual and/or automated test execution.

Maveryx is an automated software testing tool that provides functional UI, regression, data-driven and codeless testing capabilities for a wide range of desktop and Web technologies. Its innovative and intelligent technology inspects the application’s UI at runtime as a senior tester does. With Maveryx, there is no need for code instrumentation, GUI capture, maps and object repositories.

From Visual Test Design for Manual Tests…

With Yest Augmented by Maveryx, the functional manual testers concentrate on creating graphical workflows and define actions including the expected results for the various cases to be tested. Yest then generates test scenarios from these workflows and the stored information, which can be executed directly manually.

…to Automated On-The-Fly Test Execution

But can they also be automated and, if so, what results can we expect? This is where Maveryx comes into play. All you have to do is to provide a detailed instruction of manual steps in Yest. Yest Augmented by Maveryx recognizes these instructions for manual execution (for example “Click on Submit”) and executes them automatically.

If a step is not interpretable or fails, Yest Augmented by Maveryx stops and waits for manual input. You may then abort the test or do the necessary steps manually and continue. The execution results are reported in Yest Augmented by Maveryx or in the test management you read the test cases from.

Use Cases

Yest Augmented by Maveryx serves in various situations:

1. Hardening the test procedure : Manual tester may execute their tests without having to code a single line of code. This provides them with rapid feedback on the quality of their tests. No more tedious analyses and hot disputes about whose fault it was.
2. Executing manual tests directly from Xray or MS Excel : During official test runs, manual testers can call upon Yest Augmented by Maveryx to do part of the work. In fact, it is possible to import tests from Xray or MS Excel and execute them even without having used Yest Augmented by Maveryx for test design.

As functional manual testers embark on their automation journey, they will gain valuable insights into creating robust, maintainable test suites that stand the test of time.

Conclusion

This blog aims to guide manual functional testers through the complicated process of developing functional tests and harnessing the transformative power of automation. In an age where software development demands agility and speed, understanding the intricacies of automated functional testing is critical for testers looking to optimize their workflows.

Clarity, modularity and maintainability are key success factors for a successful transition from manual to automated test execution. Visual test design and model-based test case generation pave the way to structured tests and complete coverage. With the right tool support, it is possible to carry out these tests automatically without having to write a single line of code. Yest Augmented by Maveryx provides this support. Contact us to learn more at info@maveryx.com or contact@smartesting.com

[1] Khankhoje, Rohit. (2023). Revealing the Foundations: The Strategic Influence of Test Design in Automation. International Journal of Computer Science and Information Technology. 15. 10.5121/ijcsit.2023.15604.

Authors

Alfonso Nocella, Co-founder and Sr. Software Engineer at Maveryx 

Alfonso led the design and development of some core components of the Maveryx automated testing tool. He collaborated in some astrophysics IT research projects with the University of Napoli Federico II and the Italian national astrophysics research institute (INAF). Over the decades, Alfonso worked on many industrial and research projects in different business fields and partnerships. Also, he was a speaker at several conferences and universities.

Today, Alfonso supports critical QA projects of some Maveryx customers in the defense and public health fields. Besides, he is a test automation trainer, and he takes care of the communication and the technical marketing of Maveryx.

Anne Kramer, Trainer and Global CSM at Smartesting

Anne Kramer first came into contact with model-based test design in 2006. Since then she has been burning for the topic. Among other things, she was co-author of the “ISTQB FL Model-Based Tester” curriculum and lead author of the English-language textbook “Model-Based Testing Essentials” (Wiley, 2016).

After many years of working as a process consultant, project manager and trainer, Anne joined the French tool manufacturer Smartesting in April 2022. Today, she is fully dedicated to using models for testing purposes. This includes visual test design and, more recently, generative AI.

Maveryx is an Exhibitor at EuroSTAR 2024, join us in Stockholm.

End-to-end testing is one of the most effective ways software teams can understand the customer experience. Unlike unit or component testing, which focus on individual pieces of the application, E2E tests seek to understand product quality as an integrated journey. In many ways, end-to-end testing exemplifies the expanding role of software testing in a DevOps world: the crucial connection between how software is built and how it is used. When development teams understand how their changes will impact their end users, they’re better able to deliver value to those customers. When automated effectively, end-to-end testing provides this connection quickly enough to support continuous delivery.

But the shift to digital-first experiences means that end-to-end testing needs to evolve and expand, running contrary to established testing best practices. Even a simple user journey, such as the one outlined below, likely involves multiple third-party APIs and services as well as email touchpoints and personalized offers or recommendations. Development teams must build seamless user experiences that make a complex customer journey feel simple. Expanding the definition of end-to-end testing ensures they can do so successfully. But with traditional testing frameworks, complicated automated testing meant a high risk for broken tests, extra maintenance, and inaccurate results, which ultimately slowed down development pipelines. Quality teams instead opted for shorter, simpler end-to-end tests that were less likely to break as the product evolved. But the holistic view provided by true end-to-end testing is extremely valuable – if quality teams have the tools to manage them.

New Customer Journeys Demand Broader End-to-End Testing

The image above includes an example of an end-to-end test for an ecommerce website. Despite this being a fairly simple – and common – transaction, an automated test needs to cover a marketing email, a coupon code, and an invoice email with a PDF attachment. But the story doesn’t end there: it’s extremely likely that the checkout test step includes an API for a payment service like Square or Afterpay. It’s also likely that coupon codes are personalized for customers, given that loyalty programs with customized rewards are proven to increase consumer spending.

Skip these steps, and there is a real risk to revenue. If a marketing email fails to accurately show a customer’s coupon code, conversion rates will suffer, impacting sales and potentially churning previously loyal customers. Managing this type of comprehensive test is essential for supporting quality customer experiences.

The Challenges of Comprehensive End-to-End Testing

Though the above end-to-end test is critical for understanding the user experience and how each change will impact it, such tests pose several challenges for developers and software testers. First, maintaining such an extensive automated test with scripted test automation frameworks is likely to consume a significant amount of a testing team’s time and effort, which has a serious impact on an organization’s ability to accelerate product velocity. Since additional test steps increase the risk of a test breaking, most quality teams avoid creating longer end-to-end tests in order to reduce the burden of test maintenance. But what they gain in reliability, they lose in test coverage.

Second, comprehensive automated tests often require longer investigations into failures. Combing through a long list of test steps to identify the specific cause of a test failure can take valuable hours, a luxury development teams don’t have as delivery cycles shorten. Considering that 44% of developers say that investigating failed tests is a significant pain point, quality teams must have effective strategies in place to triage comprehensive end-to-end tests when necessary.

Maintaining More Complex End-to-End Tests

An end-to-end test covering email, API, and non-functional test steps is highly susceptible to any product changes, but advances in AI and machine learning have reduced the amount of time and effort needed to maintain automated tests, making it possible for quality teams to manage comprehensive end-to-end tests. Using unique identifying elements across an application’s UI, including shadow DOM components, intelligent test automation solutions can detect product changes and update end-to-end tests accordingly.

Automating end-to-end test maintenance not only ensures that test maintenance is less labor-intensive, but also allows more team members to contribute. For example, manual testers can more easily collaborate on E2E tests that contain integrated API tests, ensuring that comprehensive end-to-end tests capture the full user journey and accurately assess quality.

Reporting on End-to-End Testing

Even when end-to-end tests are maintained, identifying the root cause of an error can be time-consuming, causing delays and disruptions in the later stages of the SDLC. Rapid results that support fast bug resolution is critical for delivering exceptional user experiences at the speed of DevOps.

Advances in cloud-based testing and the availability of SaaS test automation tools are making it easier to scale and maintain comprehensive end-to-end testing strategies. Cloud-based runs give in-depth insights that support continuous improvement, and can be run on a schedule or as part of a CI/CD pipeline. Flexible execution options make it possible to routinely and reliably run comprehensive end-to-end tests without slowing development. But perhaps even more importantly, integrating end-to-end test automation into existing development workflows allows developers to quickly act on end-to-end test results.

Building workflows that surface comprehensive end-to-end test results in a digestible way supercharges their value. Sharing test results as Jira tickets, complete with screenshots of the point of failure, DOM snapshots, and performance logs, is ideal for triaging comprehensive end-to-end tests since developers can easily identify what test step caused the failure. The time from failure to fix becomes much shorter, making comprehensive end-to-end tests highly actionable.

The Future of End-to-End Testing

comprehensive end-to-end tests are often considered too time-consuming to provide real value to development teams. But their ability to ensure quality of the perspective of the customer is invaluable, even essential, in a time where every business is competing on their digital customer experience. Overcoming test maintenance, execution, and investigation obstacles to comprehensive end-to-end tests gives development organizations a powerful tool for understanding how changes will impact their users. And with the right test automation solution, end-to-end testing becomes an adaptable process that can continuously evolve to match real customer needs. A few examples include automated accessibility checks, integrated API tests, shadow DOM components, and cross browser testing. No matter what your customers need, comprehensive end-to-end tests will help your team deliver exceptional user experiences. 

Author

Bridget Hughes., Content Marketing Manager at mabl

Bridget is the Content Marketing Manager at mabl, the unified test automation platform for delivering modern software quality. She’s dedicated to helping quality teams expand testing and improve product quality through educational blogs, articles, and the occasional software testing meme.

Mabl is an Exhibitor at EuroSTAR 2024, join us in Stockholm.