Business Management

Mudit Kumar (APAC) - Software Testing Trends in APAC

Testing as a practice has grown in complexity and significance from the time software industry was in adolescence in early 1990s - I would name pre-Google era as Testing 1.0, post Google until 2006 as Testing 2.0 and post that as Testing 3.0.

In Testing 1.0 period, testing was conducted in a simple, adhoc way without any specific processes or methodologies. Testing was part of the debugging process and not really an independent activity per se. Success in testing was mainly achieved by an individual's brilliance and not due to any set standard or process. The testing techniques were adhoc without any structured documentation. The financial model was predominantly input based on traditional single service & domain.

Then, Google came along with many other technology companies with niche applications and products. This era showed significant improvement in test processes and methodologies. Dot com companies created stress on the network and therefore required agility on the network provider's part to deliver services on time despite growing complexity. This era also marked recessionary spikes that forced many companies to reduce their R&D spend but at the same time sustain the current business. This required testing to become smarter.

Testing 2.0 was an era of evaluating an attribute or capability of the system and determining that it met required results. Testing processes were aligned to SDLC. Infrastructure started becoming more planned and dedicated for testing. Testing as an activity started getting more focused and evolved as an independent practice. New models like Testing COE consolidated disparate testing activities under one roof, bringing optimization in processes, tools and resources.

Cost pressure continued to increase, leading to better ways to conduct testing through innovative delivery models, remote testing, offshore/onsite mix etc. Customers now had to adapt to newer technologies like 3G and newer architecture types like SOA and Cloud. With these changes, testing started seeing more focus on automation, performance and security testing. Customers explored outsourcing large testing deals comprising E2E ownership of applications, network and infrastructure. Financial models with stringent SLAs were introduced demanding better productivity of execution, reduction in cost and improvement of quality. Testing was getting more significant.

Testing 3.0 is the current era, which is extremely complex from a technology perspective. Now, the financial model is based on the solution, which is based on integration of various service lines measured in terms of the value of transformation delivered. Testing is linked to the outcome of the engagement such as defect removal efficiency (DRE). It encompasses all the risks associated with the delivery of the application in production environment.

Networks have become heterogeneous and complex, as there are integration points from various off-the-shelf products and applications. Phones have becomes smarter with more and more applications moving onto them. With these changing scenarios, customer preferences and demands have also changed; leading to changes in the way testing is done. Apart from performance and security testing, structural testing is a new methodology to uncover defects as early as in the design phase or architecture phase. Important to understand is that with the change of applications onto new architecture like SOA or Mobile or Open source, probability of defect slippage right from the architecture phase becomes higher if appropriate testing is not done. It is therefore important to have testing done in the architecture phase itself, which is called structural testing. There are tools like STAMP that are becoming market leaders in structural testing.

Another important trend clearly seen across all industries is improving test process maturity. Testing of large global organizations is done across various locations. A framework with overall governance model is required to streamline testing activities across these locations. TMMi model gives maturity mapping enabling organizations to carve out an improvement plan.

To summarize, Testing 3.0 phase has tremendous complexity in applications and architecture but to tackle it efficiently and effectively, service providers like Mahindra Satyam have offerings like structural testing, service virtualization and TMMi, on top of pure play services like automation, security testing and performance testing. Testing is now considered a revenue generation activity and not as cost reduction activity which was the case until Testing 2.0. Now, testing plays a key role in business transformation. Efficient testing improves the ROI by generating new revenue or profit by offsetting the cost.

We see similar trends APAC region, which is a growing market from a testing perspective. Customers are getting aligned to independent testing services. Industries such as manufacturing, retail, telecom and energy/utilities are the ones to look out for opportunities. Any offerings that are aligned towards asset management, infrastructure planning and monitoring are key to success. Japan hasn't yet adapted cloud services and is still sticking to legacy architecture like mainframe. China, Thailand and Malaysia are the manufacturing hubs of companies in North America. One needs to align the business model with North American companies to generate opportunities from their manufacturing hubs in Asian countries. Testing is growing and is likely to remain a focus area for years to come.

By Mudit Kumar, Assistant Vice President - Testing, Mahindra Satyam


« Kathryn Cave (Kenya) - Kenyatta and the Country's Technology Future…


Sripriya Kalyanasundaram (Global) - What Does the Future Hold for Women in IT? »


Do you think your smartphone is making you a workaholic?