Leaner and Faster Processes – Testing Times for QA providers

In the past, quality assurance and testing were considered an end-of-life-cycle activity but new market dynamics have made things way different. Yet, can we afford product recalls or sloppily-tested products?

Pratima Harigunani
New Update

Narsimha Rao


INDIA: We are living in times, where the half-life of any idea or a product or technology is rapidly decreasing. Products and technologies are getting obsolete much faster and newer ones are being launched more often. Any company planning to launch a new product or service has to walk a tightrope balancing market forces, their competitors and high expectations of customers, all the while trying to ensure that the product is of high quality. A small misstep in any of these and the backlash could potentially impinge their brand irrevocably and have dire consequences for the organization.

To give you an example, let us assume a new product-line is launched on an e-commerce website without adequate testing; or a new phone is launched worldwide with a malfunction. The cost of rectification in such scenarios is going to be prohibitive! Besides, customer’s today are spoilt for choices and are unlikely to give a second chance. Such mistakes could then have far-reaching ramifications. Quality assurance (QA) and testing thus acquire immense importance, especially when organizations are looking for ways to take their technology implementations to the market faster than ever before.

Traditionally, quality assurance and testing were considered an end-of-life-cycle activity and outsourced to an independent entity to ensure a fair and unbiased validation of the product. This is no longer true, simply because everyone is pressed for time and facing market pressures. We no longer have the luxury of an elaborate development and testing life cycle. However, that does not mean businesses can afford to release products, which have not been comprehensively tested.


AI and ML are showing great potential in identifying testing defects quickly : Narsimha Rao AI and ML are showing great potential in identifying testing defects quickly : Narsimha Rao

Conversely, we are at a stage where we need different types of testing to be accomplished in a much shorter time frame. We have seen how the automobile sector has been plagued with product recalls, and most of the times, these recalls are a result of sloppy and / or inadequate testing. The need of the hour is to have lean and effective methodologies for testing and quality assurance.

Thus, QA service providers need to focus much more on prevention of problems than before and at the same time, automate detection to reduce cycle times and cost. Efforts need to be focused on integrating testing into the entire product development life cycle instead of considering it as an isolated function. Behavior and Tester-driven development is one of the ways forward. Traditional methods of testing, where testing gets planned for and executed at the end of the life cycle, need to be examined for its relevance and in many cases testing need to be integrated completely with development. Quality control needs to become an end-to-end function in the product development life cycle.


The next important function that needs to be inducted in all QA processes is automation. Early in its evolution, the QA service delivery team at Infosys also started off testing applications and products manually. However, the number of applications increased and clients aimed to increase test coverage and lower costs at the same time.

As agile development augmented demand for more effective and faster testing, automated testing was the only way ahead. In addition, in the last four to five years, we have introduced automation to around 60 per cent of our testing activities. So much so that today at Infosys, we have successfully used robots for remote testing point of sales (POS) applications for banking customers.

Technologies such as artificial intelligence (AI) and machine learning (ML) are showing great potential in identifying testing defects quickly and eliminating human intervention. With this new development, we now also have the capability to look at how a product will perform both at the machine-level and application-data-server-level. These insights can help businesses deliver superior products.


As a strong QA service provider, we at Infosys, have always focused on reducing testing cycle time, improving effectiveness, and finding tools that can predict defects while helping to enhance performance engineering. Last but not the least, while we get busy ironing out technical aspects of product development, we need to remember that end user experience will always remain a big priority and hence, find ways to infuse customer considerations into the product development life cycle as early as possible.

This is where customer feedback on social media comes into play. Organizations need to find ways to effectively capture this information from social media and mine it in such a way so as to help product development proactively address customer concerns and ensure that these scenarios are tested effectively.

The need of the hour for all QA providers is to focus on developing leaner and faster processes so that we can help build cutting-edge technological products. Automation and usage of AI / ML techniques will play a significant role in the times to come.

(Narsimha Rao is SVP & Head-Infosys Validation Solutions. Views expressed here are of the author and CyberMedia does not necessarily endorse them.)

testing agile