Automated testing became an indispensable part of the software development business. It is also given specific business objectives to attain. Which trends will dominate automated testing in 2021 and determine who will get the biggest market share?
The rapid development of the Agile and DevOps methodology integrated automated testing into Agile (which also means accelerated) process of the code generation. One of the direct consequences of such a change was so called shift-left testing, which means moving the testing to the earlier phase of software development – it can be even gathering of the business requirements. Before “tha Agile era” testing could start only when the final effects of the developers’ work were available. It resulted in the higher costs of the maintenance and bug-fixing (because some of them were overseen in the earlier phases of the work). Not to mention huge losses of the company’s image and outflow of the customers disappointed with the product. It therefore comes as no surprise that the executive IT management surveyed in “The World Quality Report 2019-2020” recognised “contribution to business growth and business outcomes” as the most important objective for testing and QA in 2019:
Chart 1. Answers of the executive IT management concerning objectives for testing and QA in 2019:
However, testing in 2013 (when Agile started to gain in popularity) and now are two different worlds. What will contribute to increasing business results after Agile became an indispensable methodology of work in the software development business?
Automatization of the new generation
In recent years, the software testing market has been experiencing a real boom and – according to the latest analysis – it is only the beginning. As stated in the Techavio report “Software Testing Services Market by Product, Geography, and End-user – Forecast and Analysis 2020-2024”, it is expected to grow at an even faster pace – with CAGR at the level of 12,62%. Read more about business consequences of this phenomenon. The necessary element of this process is development of the automated tests, which supports changes taking place thanks to the implementation of the Agile and DevOps methodology. Currently, the tools which offer only simple functions such as recording the tests performed on the website or in the app, are insufficient. To keep up with the rapid changes in the development environment there is a need of the software which will enable for automatization of the most actions taken by the testers. Additionally, such a tool should be easy to handle for the person without any programming skills and offer the feature of quick identification of the most bug-prone parts of the code. Check out 10 characteristics that the new generation testing automation tool should offer. Automated testing ceased to be only the tool which supports testers in their work – it became an indispensable element in the end-to-end testing process.
Artificial intelligence (AI) and machine learning (ML)
Although they became very popular and catchy slogans, the newest Techavio report “Software Testing Services Market by Product, Geography, and End-user – Forecast and Analysis 2020-2024” confirms that the testing sector has just started making use the benefits which stem from this technologies. The results are encouraging, and the forecasts concerning the pace of development of the testing market (according to the TechWiz its value in 2025 could reach 190,6 billion USD) confirm its great potential. As we read in the Techavio report, implementation of the artificial intelligence in the testing process is crucial in striving for the smart testing, which means ensuring maximum effectiveness of the testing from the earliest stages of the code creation. What is important, QA specialists and testers who will be introducing the artificial intelligence in the company should receive appropriate training – it is necessary to fully benefit from the possibilities provided by this technology. Companies which will be first to discover such a need would gain a huge competitive advantage. The experts quoted in “The World Quality Report 2019-2020” indicate the complementary role of machine learning in the process of enhancing quality of the testing with the utilisation of the AI. How could it happen? As stated in the report, “As organizations grow accustomed to mechine learning, they will increasingly understand its distinction from artificial intelligence. They will see that in order for AI testing to be effective, it needs access to the right data – and machine learning can provide this, by processing records in volume, thereby helping to identify and optimize test cases”.
Big Data technology
It is another shining example how testing is intrinsically linked with the latest trends in the world of the software and new technologies, as well as with the e-commerce market. Big data is used in the development of the programmes for an automated testing (i.e. with the use of the ML described above), although the data sets also require testing and continuous monitoring. Big online shops or the software for health systems are great example here – such a huge data conglomerates require testing on each phase: introduction, processing, migration, performance and speed. Along with the progress of digitisation process, which concerns each line of business, the demand for such services will be growing.
New challenge – internet of things.
“The World Quality Report 2019-2020” mentions quick development of the testing in IoT in the following industries: car, energy and telecommunication. The latter would face a great challenge in the coming years: consequences of the 5G technology implementation. As piszą Darren Coupland (Sogeti) and Mandan Sundaraju (Capgemini) write, “This will create a new quality assurance (QA) dynamics. All these new use cases will need to be validated end to end. How so? Because the increase of internet of things (IoT) developments implicit in 5G has implications not just for IT applications, but also for the physical devices themselves. For example, it’s not just the app managing the traffic light that needs to be tested: it’s also the traffic light sensor itself”. Test coverage of the smart devices interconnected with the applications became a necessity in the times of the rapid expansion of the IoT market. As the Testing Whiz reports, in 2016 its worth was estimated for 6,4 billion USD, in 2020 – for over 20 billion USD.
Blockchain technology as a new business model
Blockchain solutions are being used increasingly with regard to the processing and protection of the data. The report “5 Blockchain Trends for 2020”, prepared by Deloitte, reflects its growing popularity and scope of use:
Illustration 1. Conclusions for the coming year with regard to use of the blockchain technology in the business.
In the coming 12 months implementation of the blockchain technology is planned to be increased in the following sectors:
⦁ technology, media and telecommunication (from 45 to 49 pct.)
⦁ energy (from 37 to 43 pct.)
⦁ industry (from 40 to 43 pct.)
⦁ life sciences and health care (from 36 to 41 pct.)
⦁ finance (from 33 to 38 pct.)
⦁ professional services (from 24 to 33 pct.)
⦁ industrial and construction services (from 32 to 33 pct.).
Testing plays crucial role in the blockchain-based products, because it monitors its security level and effectiveness of data-gathering. In these fields there is no room for error.
The increased significance of the security testing
The statistics presented in “The World Quality Report 2019-2020” confirm this trend. The survey respondents indicated that matters of security and risks are the most important for the IT strategy adapted by the company. The security testing scale in the cloud computing environment increased from 42 to 57 pct., what reflects development of the securities and the way such tests are performed, as well as adaptation of this line of automated testing to the fast pace of work in Agile methodology, because testing in the cloud enables for quicker implementation of the changes and reaction to them. In order to ensure the maximum coverage of security testing, the authors of “The World Quality Report 2019-2020” recommend introducing it on the earlier phases of the software development and carry it out with the tools for automated testing. What is even more interesting, the level of automation in security testing is surprisingly low – it discharges only 13 pct. from this kind of tests. However, if you look at the weighting attached to the particular aspects of security in the process of the software development, the increased significance of security testing seems inevitable:
Chart 3. Ranking the importance of certain aspects of secure development strategies.
Performance engineering as an essential mind-set and way of working
The time when automated testing and software development consisted only of delivering the specific solutions, and when each team had been working on its task separately from the other, can be described as the programming stone age. Nowadays the measure of success is creation of the product that all teams had been working on together. All people involved in the project (business line, project managers, developers, QA, testers) from the very beginning get acquainted with the business requirements and the needs of the end-user. They ask questions and talk about possible solutions. Thanks to such a workflow it is possible to implement many changes at an early stage, which prevents from the irreversible (or highly costly) damages that can occur after the deployment. In this way the performance engineering perfectly dovetails into the Agile objectives, because it postulates to look at the product from the meta-level and requires all teams to closely cooperate with each other. The effects are astonishing: the products that were created in this approach have a great market value in terms of the software, hardware, security, UX, effectiveness and performance.