The Katalon Blog

What is Shift Left Testing? Shift Left Meaning in DevOps

Written by Katalon Team | Jun 13, 2019 11:06:15 AM


 

As the demand to release quality software products in a short amount of time continues to accelerate, Agile and DevOps teams have to adopt new approaches to avoid falling behind. Continuous testing and shift left testing are two breakthrough solutions for them in the modern era of software delivery. In this blog, we will walk you through those two2 concepts, their benefits, and how to implement them into your software development life cycle (SDLC).

What is Shift Left Testing in DevOps?

Shift-left testing is a software testing approach that shifts all testing activities to earlier development stages rather than leaving until the very final stages. This is a proactive testing approach based on the idea that the earlier defects are identified and resolved, the lower the cost and effort required for fixing them.

Why We Need Shift Left Testing?

Even Agile teams who have broken away from the traditional Waterfall development model put testing at the last step. They believe that testing should be completely separated from development, and the product should only be tested once it’s fully developed. This approach has proven to be ineffective, especially when the software build is unstable, so testers don’t have enough time to test if they don’t start testing early.

Another downside of the traditional testing approach is that developers have to allocate a large part of their bandwidth to troubleshoot new bugs. This happens just as they are about to begin new projects, BUT those new bugs must be fixed before the app can be released to production. 

What does this mean? Instead of focusing entirely on the new project, developers have to shift their focus to fixing bugs, creating roadblocks. They have to work on 2 huge tasks at the same time, and nothing truly gets done.

Every Tester Should Know: The Bug Life Cycle and Bug Management Best Practices

In the past, the lack of technology may have not allowed for development and testing to happen continuously, but now we have a wide ecosystem of tech to support shift left testing. For example, QA teams can leverage open-source frameworks for automation such as Selenium, Appium, or Cypress to write automated tests that can be executed repeatedly and frequently on schedule, with the support of environment virtualization tools

Read More: Top 10 Appium Alternatives You Should Know

The Dev and QA teams can communicate on these bugs through popular knowledge sharing platforms such as Confluence, then the dev team can use static code analysis tools (SonarQube, ESLint, Checkstyle, etc.) to check their code based on established standards. 

Since bugs could occur at any iteration of software development, it makes sense to also incorporate testing along the way to catch them and prevent further problems as soon as possible. Combining continuous testing and shift left testing is a great strategy to improve software quality, meet users’ expectations, optimize the allocated resources, and minimize risks of bug-breaking apps from slipping into production.

In short, to embrace shift-left testing is to join the evolution of “shifting-up”:

Key Elements Of Shift Left Testing

1. Early Testing

The essence of shift left testing is moving all test activities to earlier stages in the development process, preferably at the requirements gathering and test planning stage. Key stakeholders, testers, and developers all bring their subject expertise to the table, diversifying the perspectives, giving everyone involved a comprehensive understanding of how to develop and test simultaneously. 

This alignment of vision lays a solid foundation for shift left testing: testers know how to allocate their resources to certain types of testing, which approach to use, which tools are needed, and how to scale their test activities as the product gets more developed.

Read More: 15 Types of QA Testing You Should Know

At this stage, bugs are less expensive and easier to fix. Less code was written, which means significantly less dependencies (i.e. the interactions and connections between software features) have been created. Those bugs haven’t been entrenched in the software design, so testers can troubleshoot without radically modifying the existing code. 

Another important advantage is that the development team is allowed to focus their entire bandwidth on development, then start testing only when it is time to do so. Developers no longer have to struggle to balance bug fixing with developing. 

From the developer standpoint, early testing brings confidence. Developers know that they are building the application on a stable foundation where bugs are continuously being detected and fixed. It is nice to know that they don’t have to scrap large sections of code they have spent so much time and effort writing. 

2. Collaborative Approach

Shift left testing promotes strong cooperation not only between the QA team and development team, but also other stakeholders, such as the Business Analyst or Project Managers. Traditionally, there is a clear separation between development and testing, and shift left testing fuses the 2 processes with each other, leveraging the expertise of each team to maximize the project’s efficiency, trimming the excessive processes.

With shift left testing, more key stakeholders are involved. Such diversity and variety in perspective promotes continuous communication, feedback, and effective problem solving. Developers receive suggestions on how to code test-friendly software, while testers learn about the specific areas that need attention during testing. And although it’s a non-technical factor, the variety of skills and personalities can actually promote innovation in the project.

Such an approach requires a shared understanding of the project’s objectives and quality expectations throughout the project. Adopting a testing tool with centralized communication capabilities is a great starting point to start shift left testing.

3. Automation

The essence of shift left testing is testing quickly and continuously, which couldn’t easily be achieved with human effort alone. If QA teams have decided to adopt shift left testing and continuous testing, it is wise to also invest in automation to speed up testing and increase operational efficiency

For example, regression testing - testing the application after a code change to detect bugs introduced due to those changes - is a common testing type that accounts for a large part of testing in a project, and manually retesting all features in the application is extremely time-consuming. This is because there are many repetitive but crucial test cases, such as the Login feature, that the QA team has to always test every time the code changes. The QA team can instead write a script to execute it automatically, at a predefined time, with high consistency and accuracy, without having to intervene much.

Read more: What is Automation Testing? Ultimate Guide and Best Practices 

4. Continuous Integration And Continuous Delivery (CI/CD)

Continuous Integration (CI) and Continuous Delivery (CD) are the combined practices and tools to automate the process of building, testing, and deploying software changes, making software companies more agile, flexible, and adaptable to market and business requirements. 

Here CI is the process to regularly merge code changes from multiple developers into a shared repository, whereas CD takes the CI process one step further and ensures that the software can be deployed to any environment consistently and reliably. Establishing a CI/CD pipeline is establishing a techstack that holistically supports the shift left testing approach and embodies the mindset of continuous improvement

Thanks to the CI/CD pipeline, the code is continuously integrated and tested, providing quick feedback loops to identify issues early in the development cycle. This creates a strong ripple effect, allowing for more frequent releases, faster time-to-market, and cost savings. 

Read more: Continuous Delivery vs Continuous Deployment - Everything You Need To Know

Challenges of Shift-left Testing

Shift left testing does pose several interesting challenges for any team looking to adopt it, and these challenges tend to arise in the transition phase from the Traditional model. It is a rapidly changing time for the organization, where a lot of consideration, planning, and embracing new systems happen. Below are the most common challenges we have seen QA teams encounter on their journey to adopt these modern practices:

  • Transition Planning: formulating a clear-cut plan allows both the QA team and the development team to understand what this transition entails for their work. Shift left testing impacts both of these teams, and upon adopting it both teams will work together in an intertwined fashion, so several old processes or structures should be changed to reflect this.
  • Test Planning: familiarize both teams with a collaborative  phase. No longer a linear and sequential effort conducted after the development phase, test planning now requires early involvement and establishing a continuous communication - continuous feedback loop.
  • Project Management: prepare and properly train your project managers to incorporate the shift-left testing mindset into their processes. Project managers should learn how they can align the shift left principles with the traditional project management processes with DevOps best practices. They should also foster the mindset for quality in their teams, ensuring that such cultural shifts are not too disruptive from a technical and organizational perspective.
  • Integrating Automation: Maintaining excellent software quality in shift left testing is easier with the right automation practices implemented, but a lack of automated testing expertise prevents QA teams from fully adopting automation. Test script creation demands strong programming skills, and then when the software code is updated, these scripts must also be revised - a resource-intensive process when the product scales and the number of test cases to be revised skyrocketed.
  • Support for Developers: Developers can be resistant to testing and should be prepared to add testability to their skillset. However, testability is essentially following development best practices. Testers can become advisors for developers, helping them create clean, concise, easy-to-understand code, facilitating a smooth testing process
  • Communication Silos: identify and eliminate communication silos which have always existed between the QA and Dev team due to limited interaction and a lack of a shared system of record. JIRA is a good solution for this, but JIRA is not specifically designed for software testing, so they may need a JIRA integration to turn it into a more tester-friendly tool.

Shift Left Testing FAQs

1. Why does DevOps recommend shift-left testing principles?

DevOps recommends shift-left testing principles to ensure that testing is integrated into the software development process as early as possible, which helps to detect and fix bugs and issues earlier, reducing the risk of defects being found later in the development process or in production. This results in faster delivery of high-quality software and helps to increase customer satisfaction.

Read More: DevOps Lifecycle Fully Explained

2. Is shift left testing possible in automation and performance testing?

Yes, shift-left testing is possible in automation and performance testing. Shift left testing is simply a methodology to testing where you bring any type of testing to the earlier stages, so you can literally perform any type of testing with it.

3. Is shift left and test driven development the same?

Shift-left testing and test-driven development (TDD) are indeed related concepts, but they are not the same thing. Shift-left testing is a testing methodology that involves testing early in the software development lifecycle, whereas TDD is a programming methodology that involves writing tests before writing the code. TDD can be part of a shift-left testing approach, but it is just one piece of the puzzle.

Read more: TDD vs BDD: Key Differences

4. Who started shift left testing?

The concept of shift left testing originated from the software testing community, although it is difficult to pinpoint a specific individual or group that started it.  However, several sources indicate that Larry Smith was the first to coin the term, who back in 2001 explained his concept of “shift-left testing” and how he envisioned it as a pragmatic solution to “integrate QA and development parts of a software project”.

What is Continuous Testing in DevOps?

Continuous testing uses a procedure of running tests and frequently throughout the development lifecycle. Incorporating continuous testing into your organization’s testing strategy accelerates your time-to-market and ensures that your customers receive the release quality they expect.

Additionally, continuous testing establishes a constant feedback loop from the testing team to the developers for them to fix bugs before releasing the application to production. It ensures prompt bug identification and early bug detection. The entire software development process is therefore more optimized at every stage.

Continuous testing relies quite heavily on automation, especially in large-scale testing projects, where it would be counterintuitive to manually run tests through iterative testing cycles. Automation is not a requirement of continuous testing, but it does help the process run much faster and, as such, is highly recommended.

To learn more about the benefits and challenges of continuous testing, have a look at these two articles:

Shift Left Testing vs. Continuous Testing

Have a look at this table to understand the differences in 2 approaches:

Aspect

Shift-left Testing

Continuous Testing

Definition

Testing activities are moved earlier in the software development lifecycle (SDLC).

Automated tests are executed continuously throughout the SDLC.

Goal

Identify and address defects early in the development process.

Provide immediate feedback on the quality of the software.

Techniques

Unit testing, code reviews, static analysis, early integration testing.

Automated unit tests, integration tests, functional tests, etc.

Focus

Early defect detection and prevention.

Ongoing quality assurance and feedback.

Benefits

Reduced cost of fixing issues, shorter feedback loop, faster delivery.

Maintains high confidence in software quality, faster release cycles.

Integration

Integrated into the beginning or middle stages of the SDLC.

Integral part of the continuous integration and delivery pipeline.

 

How to Apply Shift-left Testing in Continuous Testing

Now that you have understood the 2 definitions, let’s unravel these concepts to see how they fit into the big idea.

Shift-left testing is a part of continuous testing, as it employs the same process of testing at every stage and as often as possible. While continuous testing is part of the process of executing tests in the DevOps pipeline to get feedback on bugs as soon as possible, shift-left testing focuses on defined objectives. 

Therefore, if continuous testing incorporates the entirety of the DevOps pipeline (i.e. from planning, designing, developing to deploying), shift left testing takes a part of each cycle and sets testing strategies for each task in the cycle. Shift left testing enhances continuous testing and makes it a stronger weapon against bugs.

The challenges we outlined above are not to be afraid of. Below are the strategies and practices your team can adopt to integrate shift left testing and continuous testing into your strategy: 

1. Involve testers early on for planning and analysis

To truly shift left, we should involve testers early on in the planning phase. They can provide feedback on the testing feasibility of the design and provide input on potential testing scenarios. If there are any ambiguous requirements, gaps in requirements, or inconsistencies that may slow down testing, QA teams can point out, so that no unnecessary back-and-forth communication happens later down the line.

Testers can also bring their expertise to the table here by advising developers on how to code with testability in mind. For example, testers can specify when teams need to develop mocks and substitution of software modules for testing in case a certain application module is not yet ready, which are highly valuable assets for integration testing

2. Specify quality standards and establish quality culture

If we want to integrate testing into the development processes, both teams must be educated on the current testing practices, tech infrastructure, as well as each team’s technical capabilities. This requires a comprehensive approach, which involves setting clear expectations for what constitutes high-quality code and making it a priority for everyone involved in the development process.

Establishing a culture of quality requires a commitment to ongoing training and education, as well as a willingness to invest in the necessary tools and resources to support testing and quality assurance. 

3. Embrace BDD testing

In addition, you can apply shift-left testing through behavior-driven development (BDD). BDD is a versatile Agile software development approach that enhances collaboration between developers and non-technical team members such as managers.

Katalon Studio is a powerful tool for you to implement BDD into shift-left testing through Jira integration. By integrating with Jira, the non-technical department such as testers and managers can directly pull requirements from Jira to BDD files, which are already integrated in Katalon. QA testers can then write test cases and send them back to developers to run.

With this approach, managers will understand what happens throughout the software delivery process thanks to the ubiquitous language of BDD; testers will not have to wait until developers finish coding. If defects or errors occur, either the QA or development team can quickly re-run test cases and make changes accordingly.

4. Adopt automation testing

When it comes to automation testing, teams have 2 choices: build vs buy.

Testers can write automation testing scripts by automation frameworks built on Selenium to automatically execute test cases. The level of customization Selenium gives is endless. However, not all QA teams have the technical expertise to develop a full-on automation tool from scratch. A better option is to shop around for automation testing tool vendors providing testing features that suit your business needs. Such vendors have tailored the software design to current software testing best practices.

Read More: Top Good Selenium Alternatives QA Teams Should Try

A roadblock that QA teams face with automation testing tools is their fragmented nature. Each tool has its own unique features serving a specific testing type, so they have to adopt multiple tools if they want to perform many different testing types. 

For example, in regression testing, teams have to perform:

  1. API testing
  2. Web testing
  3. Mobile testing (if they have a mobile app version)
  4. Visual testing

4 different tools! It is painstaking to combine the test results from these tools into a centralized view, and there can be inconsistencies in test execution that makes automation testing seem difficult.

With a comprehensive quality management platform, QA teams can create, execute, manage, and maintain web, API, desktop, mobile, and visual testing all in 1 place. In other words, everything in the testing lifecycle. Such centralization of testing gives them the most unified testing experience and analytics, leading to better data-driven decision-making.

Whatever your choice is, always embrace automation testing. Automated tests can be run quickly, frequently, accurately, at scale, providing immediate feedback. 

How To Perform Shift Left Testing with Katalon Platform?

Katalon Platform is a comprehensive test automation solution and quality management platform for web, mobile, API, and desktop applications. It provides a unified platform for teams to plan, design, execute, and manage automated testing efforts.

Katalon Platform supports all 4 key elements of shift left testing. Early testing and continuous integration/continuous delivery can easily be achieved thanks to seamless integration with popular CI/CD tools as well as collaboration tools such as Slack, Microsoft Teams, Git, bringing transparency and enhanced communication between team members.

Katalon also offers test management capabilities. Testers can capture application elements and store them in the Object Repository, which is essentially a centralized repository to store test artifacts. These test artifacts, test cases, and data can be shared across team members and reused across projects, improving test reusability.

Curious? Witness the power of Katalon in action here:

 

Download Katalon and Start Testing (The Better Way)