10 Testing Pros Share Insights Into the Most Interesting Current Trends in Automated Software Testing
Whether you’re just getting started in security or you’re an experienced DevOps professional, testing is a mainstay of your profession. Here at Threat Stack, we have a dedicated Test Engineering Team that guides quality and allows our team to stay on top of the latest trends in automation and testing so the team can implement rigorous testing of our product. In the testing world, automation is one of the hottest trends, driven by advances in AI, machine learning, and other tools that streamline tasks that used to be manual, tedious, time consuming, and prone to error.
Leveraging automation also helps to strengthen cloud security, and prioritizing automation is a best practice for IaaS providers, helping companies achieve full-stack, multi-cloud security observability. Threat Stack’s Cloud Security Platform®, for instance, helps you proactively reduce risk, detect security incidents, and achieve continuous cloud compliance without disrupting your DevOps workflow.
So what are the most interesting trends in automated testing right now? What trends could be reshaping the way you approach testing at your organization? To find out, we reached out to a panel of testing professionals and asked them to respond to this question:
“What are the most interesting trends in automated software testing at present?”
Meet Our Panel of Testing Pros:
Read on to learn what our panel had to say.
Carl Robinson is a Senior Software Engineer in Test at Threat Stack with 15 years’ experience testing software for a variety of startups in Boston and New York. His focus is on building tools and frameworks to test high availability, scalable data platforms.
“One of the most exciting recent trends in automated testing is that we are seeing better ways to manage the testing of microservices by incorporating runtime integration tests into CI/CD pipelines.”
At Threat Stack, we are able to separate concerns for some of our testing by bringing up a portion of our infrastructure in containers to be tested when code is pushed to the services in test. This allows us to isolate individual services for testing without having to rely on a massive end-to-end test environment. Using multi-project builds, we automate the bringing-up of containers for necessary queues, databases, companion services, etc., and write our tests using ScalaTest with FeatureSpec. This allows us to use the same resources as our Dev teams, and our containerized environments are also useful for local development.
Another trend that is exciting to me is really embracing DevOps principles in testing by focusing on alerting and monitoring to bring attention to problematic code changes or fragile parts of the platform. Having quality monitoring and alerting in place for test environments, as well as production, can give an organization immediate insight into changes being made to the product and how they affect the performance and functionality of the entire system.
Kraig Martin is the Commercial Director at Storage Vault, one of Scotland’s largest self-storage companies.
“It might sound like something out of a sci-fi film, but…”
All the elements for automated testing software that can determine and actually fix problems are already with us, thanks to the growth of artificial intelligence software. With that in mind, 2019 is probably going to be the year that these improvements really hit the mainstream.
Self-healing testing software is basically an automated form of artificial intelligence that uses machine learning to recognize errors, constantly improving its algorithms the more it tests. It’s already being used in a number of plugins to find and fix problem elements in scripts. This type of technology has the ability to really take the hassle out of testing and free up software teams to concentrate on more exciting projects.
Lee Barnes has over 25 years’ experience in the Software Quality Assurance and Testing field with a focus on test automation and performance testing. As Founder and CTO of Utopia Solutions, Lee is responsible for the firm’s delivery of software quality solutions, which include process improvement, test automation, and performance management.
“Certainly, the incorporation of automated testing throughout the DevOps delivery pipeline is trending…”
However, I believe the most interesting trend is the growing realization (finally!) that automated testing is not automating THE testing (despite what AI/ML proponents might say). As an example, I’m starting to see Automation in Testing (AiT) — a term coined by Richard Bradshaw — gain more traction. That term is important because it takes the focus off automating test cases with tools and puts it where it should be — using automation principles to make testing activities more efficient or effective (or both). I believe it will help prevent the silver-bullet syndrome that plagues so many organizations that think their testing problems will be solved if they can get their test cases into a tool.
Vishnu Nallani is the Head of Innovation at Qentelli.
“2018 saw a surge in new technology trends, and test automation solutions also saw a lot of transformation…”
As we move forward, we’re seeing a shift from just being focused on automated functional UI test cases to a holistic approach to enable true continuous testing.
There was a surge in AI- and Machine Learning-based test automation tools in late 2018, and we will continue to see that growth. Tools that have the ability to autonomously navigate through user journeys and also look at performance and security issues while they do that will continue to rule the test automation scene.
There has also been an increase in the adoption of Behavior-Driven Development (BDD) and Acceptance Test Driven Development (ATDD) kinds of frameworks, such as Cucumber, to help in the continuous testing area. More enterprises are now comfortable using open source test automation solutions such as Selenium and Appium with the trend only going up as automation continues to innovate. Many companies have built their business model around a more user-friendly adoption of these open source solutions.
Integrating test automation as part of a CI/CD pipeline will no longer be a nice to have feature. As companies adopt DevOps practices, test automation will become an integral part of CI/CD.
Test automation solutions will follow the trend of supporting visual testing and increase the ability to automate more applications and eventually mature to become RPA (Robotic Process Automation) kinds of solutions. We have already seen this happen with tools like Automation Anywhere.
Test automation reporting will become much more important moving ahead, as test scripts become part of the CI/CD pipeline. Reporting frameworks to save and show the history of test results along with analytics on test execution trends and most common issues will become more critical.
As technologies evolve, especially in the AI, blockchain, and IoT space, we will see more test automation solutions created to support testing in these areas.
Sia Mohajer is the Founder of Privacy Australia. For the last three years, he’s been helping people around the world reclaim digital sovereignty. Data is the new oil. Don’t give yours away.
“We’ve been working a lot in automated testing for VPNs and other security protocols, so…”
Some of these are more specific than others:
- CI/CD and DevOps will play a very important role.
- Robotic Process Automation will be implemented by more and more companies.
- Cloud-based testing platforms will get more preference, just to save time on multi-browser testing (Sauce Labs and Browser Stack). Testers will not waste much time on building frameworks; the focus will be more on automating applications on already-built, robust, open source frameworks.
- Docker and other similar tools will be used for the entire cycle to solve basic Dev, QA, UAT, and production issues.
- Companies will be preferring more SDET profiles (Software Development Engineer in Test).
- To test advanced technology-based apps, companies will only prefer candidates with the development knowledge of the same (big data testing or blockchain).
Bartek started his testing journey with video game testing, then moved forward to a FinTech company to finally become a Test Lead for Zety. He is responsible for every deployment, creating Test Teams and implementing Automated Testing Solutions. He’s a video and board games nerd.
“Automated testing allows testers to…”
Quickly and effectively check whether a specific app/ feature or web page functions (smoke tests) and allows faster regression testing. The two trends that will work for me are:
- Continuous integration solutions. This means that your deployment does not require disabling the functionality of the tested feature and allows you to test solutions on clients (data) in batches.
- Big data testing. A great solution when you work on vast data, it checks connections and dependencies between certain elements of given data. Big data testing mainly revolves around monitoring the performance and functionality of big data systems. Continuous automated testing of this type is necessary to ensure that most modern brands maintain the correct information flow, upon which their success is built. Terabytes of data checked for consistency between processing stages, monitoring the resilience and speed of hundreds of processing nodes, or ensuring the stable performance of big data base machine learning algorithms are just a few of the aspects of big data testing.
- FinTech AutoTesting in finance technology to test whether payments and invoices or renewals solutions function together properly. Automated testing has shortened the amount of time required for testing and with limited input required form the tester.
Khan is a QA and development professional with over six years’ experience delivering high performance technology solutions. As Jibestream’s QA lead, Khan leverages his passion for DevOps and process automation to manage and optimize the cloud infrastructure for production and development environments.
“The most interesting trend in automated testing we’re seeing is…”
The rise of “automated testing as a service” platforms that leverage self-learning/healing features powered by artificial intelligence. Many of these platforms have designed very clever ways of using drag-and-drop processes to write codeless tests. Once organizations have had time to do more experimentation and feasibility testing with these platforms, I think we’ll see a greater uptake of them in both small and large organizations.
Neil Price-Jones, MSc., MBA, CMST, CMSQ is the President of NVP Software Solutions.
“We are seeing two major trends …”
On the one hand, we have separate test tools and methodologies for almost every platform or SDLC, and many are getting invented or configured in-house. Tools are being created and marketed with specific ends in mind.
On the other hand, we see many of our clients desiring a consolidated status report and the ability to control and report on the testing from a single interface. So, some vendors are creating tools that interface with almost anything in the market (or they are willing to create an API to serve that need) in order to provide this consolidated control and report.
With regard to the SDLC changes, we are also seeing a desire to incorporate automated testing as an integral step in the SDLC (no longer a separate step for some of our clients). So the testing report has to be consolidated with the build and deploy reports. I realize that some organizations are a long way down this path with DevOps and the ability to build and deploy many times per minute. However, not all of the industry is that advanced and many are still wrestling with less automation.
Dhaval Sarvaiya is a co-founder of Intelivita, a software development agency based in the UK and India. He helps entrepreneurs build startup MVPs and larger companies achieve the goal of digital transformation.
“The most interesting trend in automated testing is…”
AI Test Automation Assistance.
AI is here to revolutionize testing and test automation with the aid of intelligence demonstrated by machines. In 2019, software testing will be hugely impacted by artificial intelligence and achieve explosive growth. In software testing operations, the focal point of next-generation automation testing operands will be led by artificial intelligence. The AI process automation will transform the QA operation, enabling a machine to imitate intelligent human behavior. Predictive analysis of AI test automation assistance will play a vital role in achieving test automation greatness.
How AI Test Automation assistance will influence software testing:
- Dynamically write new test cases based on user interactions by data mining logs and behavior
- Help reduce effort by optimizing test cases
- Provide deep insights into test case patterns
- Enable risk-based decision making
- Enable the root cause of issues to be discovered and reduce test cycle times
- Use app log files to generate a user experience model and provide deep, detailed insight
T.J. Maher is a Software Engineer in Test at Threat Stack developing Threat Stack’s UI and API test automation. T.J. is active in the Boston QA community as the Meetup Organizer of the Ministry of Testing – Boston. He has published stories of his automation journey in TechBeacon, and is a contributor to the book Continuous Testing for DevOps Professionals.
“Today’s biggest trend in automation? Education for all!”
It can be quite challenging finding up-to-date information when trying to switch from manual testing to automation development. Prominent members of the testing community have stepped up to meet that challenge: Angie Jones, Lisa Crispin, and Richard Bradshaw, for example.
Angie Jones, an engaging writer and speaker in the automation space, has recruited thought leaders in the testing industry to create courses for Test Automation University, a free resource for automation developers looking to upskill in automation, performance and load testing, API testing, and scaling tests using Docker.
Lisa Crispin, co-author of Agile Testing: A Practical Guide for Testers and Agile Teams, has been recruited by Mabl to talk about what they have dubbed DevTestOps, giving many webinars helping testers model their test automation strategy, shift to continuous delivery, and analyze UI Test Failures.
Richard Bradshaw, head of the UK-based worldwide software testing networking organization, Ministry of Testing, has created free courses about Automation In Testing where a tester can learn about programming basics and learn tools such as Selenium WebDriver for browser testing and Rest Assured to handle API testing. The Ministry of Testing sponsors a series of 30 Days of Testing challenges so testers can start exploring topics such as API Testing, E-Commerce Testing, and Accessibility Testing.
A Final Word . . .
As you see, our experts have many ideas of where automated testing is trending in 2019. From AI Test Automation and advances in tooling to automated testing as a service and adding automation to your CI/CD pipeline, there are many different ways to approach automation depending on your product, development practices, and goals.
However, as T.J. Maher points out, technology is ever changing, and the implementation of the current trends in automated testing all are affected by the education, professional development, and opportunities that are offered.