Paul Lachance, Chief Industry Strategist & Technologist, Software Equity Group.
When I started building software, I couldn’t have predicted what the technology landscape would look like 10 years later.
I’ve personally navigated the transition from DOS to Windows to cloud computing, and now I’m watching as AI is changing how software is developed, delivered and used.
With the pace of technological change accelerating, it’s becoming difficult to predict what the landscape will look like even a few years from now. Yet enterprise software companies must continue building products that customers expect to rely on for years, sometimes a decade or more.
I’ve experienced this challenge firsthand. I founded BigfootCMMS, a maintenance management software company that evolved through multiple technology generations. Eventually, the company was acquired, and the technology continued evolving within larger organizations, ultimately becoming part of Siemens.
Seeing something I helped create continue growing and serving customers long after my involvement reinforced an important lesson: The decisions that make a software product successful today aren’t always the ones that will keep it relevant 10 years from now.
Technology will change, as will economic conditions, competitive pressures and customer expectations. Software companies have to make ongoing decisions about where their customers and industries are heading, which changes are worth pursuing and where to invest their limited resources.
Understand where your customers are heading.
The problems customers need to solve often remain remarkably consistent over time, even as the technology available to solve them changes.
Consider maintenance management software. For decades, manufacturers have needed to protect their physical assets, extend equipment life and minimize costly downtime. Traditionally, software supported those goals through work orders, preventive maintenance schedules and equipment records.
Today, those same companies can monitor equipment in real time, predict failures and use AI to diagnose problems and recommend corrective actions. The underlying business objective hasn’t changed, but the tools and intelligence available to achieve it have become far more sophisticated.
At the same time, the manufacturing industry itself is changing. Manufacturers are adapting to shorter production runs, greater product customization and more complex production requirements. That changes how they schedule production, manage equipment and coordinate operations, creating more demands on the software they rely on.
The same principle applies across healthcare, education, government and almost every other industry. Building software that remains relevant requires anticipating changes in how customers operate, not just keeping pace with technology.
Build a foundation that can evolve.
I’ve seen enough technology transitions to know that today’s architecture probably won’t be tomorrow’s preferred approach. But that doesn’t mean software companies should rebuild their products whenever something new emerges.
Modular components, accessible data, well-designed APIs and flexible data architectures make it easier to extend capabilities, integrate new technologies and replace portions of a platform without starting over.
AI creates another opportunity. By combining operational data, domain expertise and information from interconnected systems, applications can interpret changing conditions and recommend actions.
A maintenance platform, for example, could use equipment sensor data, maintenance history, production schedules and parts availability to anticipate problems and recommend corrective action as new information becomes available.
Technical debt also deserves attention. Prioritizing new features while postponing architectural improvements eventually makes enhancements more difficult and expensive.
Established software companies may already possess decades of operational data, embedded workflows, customer relationships and specialized industry knowledge that AI-native startups can’t easily re-create. The challenge is incorporating new technologies while preserving those valuable capabilities.
Know when to adapt and when to stay focused.
Here’s something I learned the hard way while building my company: You can become so focused on keeping customers happy that you lose sight of what you’re trying to build.
One mistake I made was trying to be too many things to too many customers. When an important customer requested a specialized capability, it was tempting to say yes. Sometimes, they were even willing to pay custom development fees, making the opportunity especially attractive.
Every custom feature consumed development resources, introduced additional complexity and created functionality we might need to maintain for years. Too many exceptions could pull us away from the broader market we were best positioned to serve.
Over time, I learned the value of building reusable, configurable capabilities that could accommodate different customer requirements without creating entirely separate solutions.
That discipline becomes even more important as AI accelerates development. Just because you can build something quickly doesn’t mean you should.
I’d rather see a company invest in one AI capability that meaningfully improves a customer’s workflow than introduce five features simply because competitors are doing so.
The decisions you don’t make can be just as important as those you do.
Build for change, not for a specific technology.
One of the challenges of living through technological revolutions that are happening faster and faster is that it’s becoming difficult to predict what software will look like even a few years from now, let alone 10. I doubt anyone can say with confidence.
But companies can prepare by understanding the forces shaping their customers’ industries, building adaptable technology and making disciplined decisions about where to invest.
Those decisions also shape enterprise value. A software company that can evolve with its customers without continually rebuilding its platform or disproportionately increasing its costs is better positioned for durable growth.
The technology you build today may look very different in 10 years. But if you build a company capable of adapting, what you create can continue delivering value to customers long after the technology itself has changed.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?