Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

How to keep the next technology glitch from wreaking havoc on your business and your customers

Why do these system failures continue to occur? And what can we do to lessen their impact, if not eliminate them? Let me propose some answers.

How to keep the next technology glitch

[Mookda / Adobe Stock]

Profile photo of Amit Walia

CEO · Informatica

When computer systems fail, businesses usually recover quickly, but not always. And when glitches turn into wider outages, our technology-driven lifestyles can be agonizingly disrupted. 

In July, we all got a harsh reminder of how impactful a single software bug can be when a routine configuration update from the security vendor CrowdStrike crashed Microsoft’s widely used Windows operating system. Airlines, banks, hospitals, package delivery, coffee shops, and myriad other commercial and public services were delayed, postponed, or inaccessible. One media commentator called it “the world’s worst IT outage.”

By one estimate, that mishap cost businesses $5.4 billion. And that doesn’t include the inconveniences many people suffered as they were forced to reschedule or reroute everyday plans. 

Like other business leaders, I’m on high alert every time I learn about a software bug that has the potential to cascade. During the CrowdStrike incident, one of the first things our team did was reach out to our software vendors and partners to ensure that neither we nor our business customers would experience downstream impact. Fortunately, we avoided major issues. 

But there’s always a next time. So it’s time to ask some basic questions: Why do these system failures continue to occur? And what can we do to lessen their impact, if not eliminate them? Let me propose some answers.

OVERCOME COMPLEXITY 

One contributing factor is that, while information technology (IT) continues to advance and enable business innovation, it’s also getting more complex. There are millions of lines of software code in computers and billions of devices in today’s spiderweb of global connectivity. It’s not unusual for one of our enterprise customers to have hundreds of interconnected applications and cloud services. 

Observing best practices in software development, IT management, and business resilience can go a long way toward minimizing the downtime associated with tech glitches. That includes things like software development lifecycle (SDLC) management, rigorous testing, governance, and even bug bounty programs.

Of course, software engineers and architects are always looking to improve reliability and availability. One of the best ways to do that is to replace complexity with ease of use. For example, application programming interfaces (APIs), no-code cloud connectors, and automation can all help establish more seamless and reliable computer systems and infrastructure. 

And, crucially: As more organizations tap into cloud services, a multi-cloud model can eliminate single points of failure. 

PRACTICE CAUTIOUS OPTIMISM 

There are other signs of hope. The next big thing in system automation involves employing machine learning and generative artificial intelligence (GenAI) to reduce the complexity of today’s infrastructure. 

GenAI-enabled copilots can help optimize and even repair software and tech infrastructure. They do this by enabling observability, self-healing, and orchestration of data pipelines and computing resources to keep them running smoothly. To give you some idea of how they work, Informatica’s GenAI copilot utilizes metadata from 86 trillion transactions per month to automate thousands of operational tasks in seconds. 

Most organizations are still in the early stages of implementing AI-enabled automation, multi-cloud configurations, and other methods of boosting resilience like geographically distributed data. All of these are examples of the IT modernization that is poised to accelerate as more companies adopt and integrate AI capabilities. As businesses take these modernization steps, they should become less susceptible to old-fashioned software glitches. 

New technologies alone won’t solve the problems caused by faulty code, software misconfigurations, or that other “gotcha”—human error. Businesses also need modern data architectures that weave together these new technologies with their existing infrastructure. Here too, AI adoption is an opportune time to update architectures with an eye on digital resiliency, as I know many CIOs and CTOs are doing. 

Of course, none of this will matter if there are security weaknesses in your computing infrastructure. Even the slightest software bug can be a way in for hackers. So, best-in-class IT management and cybersecurity go hand in hand. 

BE PROACTIVE—AND REACTIVE

The July CrowdStrike update issue is behind us. But as I noted, there’s always a next time, which is why my team has processes in place to respond quickly. Preparedness means being proactive in identifying and addressing potential risks in advance, as well as reactive in how we deal with trouble. Because we have an incident response plan in place, everyone knows what to do when the next glitch threatens to grow into something more serious. 

I also recommend that your service-level agreements include timely notification in the event of a software issue to ensure that vendor support is there when you need it. 

Even if you do everything right, it may be impossible to avoid every blip in our always-on world. But tech modernization and automation can help. They can simplify digital complexity and—ideally—minimize the damage from the next Next Time.