This summer, failure to turn around Volkswagen’s software unit cost CEO Herbert Diess his job. A household name, Volkswagen risked losing its competitive advantage to China, Japan and other markets by not modernizing, alongside the reputational damage of producing cars that can’t match competitors’ software standards. Diess became perhaps the most high-profile executive to pay the price for failing to modernize an organization. He won’t be the last.
The automotive industry is just one of many that have become software-centric. Retail is now online, not brick-and-mortar. Banking is done on apps, not over the counter. Executives need to balance a different set of priorities from those of their predecessors if they are to ride the pending economic storm.
Diess’ departure wasn’t due to sheer ignorance, nor was it for lack of recognition of the importance of software development to the auto industry’s future. Upon his appointment, Diess seemed to recognize the need for change. He said that VW “needs a software culture.” What that means in practice, however, is hard to define, and often the perception of those at the top is not mirrored by experience.
A solid software foundation
Most software issues aren’t due to an inability to build; they are typically caused by a problem in the code that leads to a program crashing or producing false results. In other words, issues are caused by an inability to test. While some application failures can be quickly resolved, many can cost companies a significant amount of time — losing that competitive advantage and revenue, and ultimately, damaging to reputation.
Intelligent Security Summit On-Demand
Learn the critical role of AI & ML in cybersecurity and industry specific case studies. Watch on-demand sessions today.
As companies accelerate their digital transformation to gain a competitive advantage, the pace of software development continues to increase at an overwhelming rate. This puts enormous pressure on IT teams to quickly develop and release software, often with limited resources. However, a critical step in releasing new software is deploying the right testing capabilities to find and fix any errors beforehand. Despite this, risks are taken and corners are cut. This is where CEOs who are not immersed in a software-centric culture fail to understand the steps needed to digitally transform a business.
You need to build from a solid foundation. To use a soccer metaphor, you need to build from the back. Quality assurance and testing teams are like the goalkeepers, the safe pairs of hands that prevent falling behind and allow others in the business to focus on scoring. Too often, executives come into a business and focus on rushing out ill-developed software to catch up. You can sign as many star strikers as you want, but if you have a leaky defense, you’ll suffer in the long term.
The crucial test
The understanding of where software issues lie can vary. In banking, outages are often blamed on the complex IT systems of High Street banks, which have been built on incremental upgrades of legacy systems. Less accountability is taken for substandard software testing, despite the huge consequences for customers, reputational damage and hefty regulatory fines.
Software testing is critical to identifying defects and bugs in the code. It also enables you to see if the program matches the expected requirements for function, compatibility and design. Historically, software testing was done manually, with testers playing the role of end users and using the application’s features to ensure correct behavior. But as software development becomes more rapid and complex, manual testing can no longer keep up. Companies are attempting to transition to automated testing to manage more sophisticated requirements.
With a major global developer-skills shortage, and companies making layoffs, there are severe testing bottlenecks, increased costs, and delays to project delivery times as development teams try to upskill manual testers, re-position remaining talent and lean on their existing developers for support. At the same time, companies are navigating tight deadlines to deliver quality at speed and meet new customer demands. This all culminates in an underwhelming 15% of software testing being successfully automated today, despite the rapid rate of digital transformation.
Avoiding the reckless path
However, despite the critical nature of software testing, a significant number of companies still rely on manual processes to test their software. This means they’re vulnerable to risk, as 70-75% of outages are caused by human error. Manual testing is also resource-intensive and requires specialist skills, which many companies lack. So, instead, they cut corners and release software that hasn’t been properly tested.
CEOs and testers are not singing from the same hymn sheet when it comes to testing, and a “fix it in post” mentality is creeping dangerously into institutions that cannot afford to take risks. On the current trajectory, more and more companies will struggle with failures and outages, which could cause them significant financial and reputational damage. It will also see them fall behind competitors, and for decision-makers, that seldom spells good news for their own prospects.
Staying on the current path is reckless, but worse, easily avoided. Businesses need to tighten up their software testing practices by considering a different approach and embracing solutions like test automation systems that don’t require coding skills. This opens up the barrier to entry for software testing, and ensures that companies can access a broader range of talent to manage the process. Only then can the testing process be improved, challenges overcome and attitudes changed.
Christian Brink Frederiksen is cofounder and CEO at Leapwork.
Welcome to the VentureBeat community!
DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own!
Read More From DataDecisionMakers