• Home
  • /
  • Software
  • /
  • Get on the software testing speedway and don’t look back

Get on the software testing speedway and don’t look back

Continuous testing is no longer regarded as a speed bump in the DevOps cycle, as it’s increasingly being shown to accelerate application delivery and quality.
Greg Harrowsmith
By Greg Harrowsmith, Pre-sales CA Southern Africa
Johannesburg, 07 Apr 2021

The continuous release of software applications that consistently deliver world-class customer experiences are crucial to survival and competitive-edge in the application economy. To achieve continuous release requires collaboration across each phase of the software development lifecycle.

As the application economy drives companies to rollout applications more quickly, companies are seeing testing in a new light. It was once considered a speed bump on the DevOps fast-track but new tools and testing methodologies have emerged that have brought testing up to speed.

All businesses seek to improve the quality and speed of application delivery while at the same time reducing costs. To achieve these benefits, it is necessary to examine and possibly reset your approach to one that:

  • Allows you to design requirements with clarity so potential defects can be addressed with a limited time incursion.
  • Helps you to test more efficiently to run the smallest number of tests possible to cover 100% of the testing criteria.
  • Engenders automation in test case design, test generation and change implementation.

The challenges and how to overcome them

The biggest issue is usually the number of manual tests that slow down the entire process. Defining requirements, designing test cases and other manual testing activities can cause businesses to move more slowly and can be a complete roadblock to plans for ramping up application quality and delivery.

Implementing the speed of Agile accelerates the rate of test data creation.

Most business have experienced waits of days or weeks by test teams waiting to receive the right test data, or spending long hours manually developing it themselves. Moving on from this tiresome process relies on acquiring production data and testing at the speed of agile.

In this regard, explore new technologies that:

  • Support more data sources, including unstructured data, along with information in NoSQL environments and mainframes.
  • Mass-produce test data automatically and quickly using sophisticated synthetic data generation that simulates the variability and formats of production data.
  • Mask data the right way to get the most out of production data when needed (without sacrificing security or violating compliance protocols).
  • Mitigate problems and achieve real cost savings.
  • Enable extensive and flexible synthetic data creation.
  • Have access to their own copy of test data via virtual test data creation.
  • Can ramp up their data creation speed to match the pace of the agile environment.

Achieving the foregoing makes finding, improving, protecting and delivering test data easier. This gives testing teams greater ROI from tools during this phase of the software development lifecycle.

What are the other roadblocks to agile turnaround?

Ambiguous requirements translate into poor outcomes and big risks. When requirements are not clear, you open the door to an array of problems, including, but not exclusive to, defective software; delays and clashes between the business and IT, to name just a few.

Today, most testing is still done manually, but these methods are thankfully becoming a remnant of the past as they don’t provide enough coverage – manual testing covers only 10% to 20% of functionality.

As the application economy drives companies to rollout applications more quickly, companies are seeing testing in a new light.

Test early, test often, should be the mantra of any organisation seeking to stay competitive in the application economy. In the modern software factory, teams should be offered the ability to run tests to ensure application quality, regardless of their place in the development cycle. As you bridge and broker enterprise coverage at scale from the Test Centre of Excellence, performance engineer the applications in the cloud for continuous quality.

The statistic that only 16% of test activities are automated on average should at least push you to evaluate the risk levels of releases.

It’s important to remember that today testing can’t be an isolated workflow disconnected from application release processes. Quality needs to be front and centre of everyone’s work approach and must become a shared responsibility across development, testing, release and operations teams. These teams need to be able to work together effectively and efficiently to deliver high-quality work, on time.

Unfortunately, much of testing still relies on manual methods − hence the above statistic. Moreover, testing is often siloed across an organisation − disconnected people and tools − leading to inconsistent visibility into testing progress and results. This lack of collaboration, visibility and feedback across release processes can create unnecessary delays and risk with serious negative business consequences, including:

  • Missed deadlines: Every day a release is late is lost revenue that cannot be recovered and of course, lost competitive-edge.
  • Lost time: Too much manual labour and unplanned reworks required. All leading to inefficiency in the pipeline with wasted hours that could have been spent on innovation and enhancing market-edge.
  • Disappointing quality: In a world of choice, today’s tech-savvy consumers abandon bad service providers very quickly.

Focusing on producing the highest quality apps is critical. With the right approach, businesses can bring in more automation, and start intelligently testing apps whilst achieving a 30% reduction in release risk with a concomitant improvement in pipeline efficiency.