Our SaaS and Software expert, Vijay Raina, is a specialist in enterprise SaaS technology and tools. He also provides thought-leadership in software design and architecture. Today, he joins us to dissect a seismic shift happening across the industry, driven by artificial intelligence. We’ll explore how AI is fundamentally rewriting the rules for how software is built, sold, and valued, moving beyond traditional IT budgets to target the vast expanse of the global labor market. Our discussion will cover the evolution of software from a simple record-keeper to an active “digital worker,” the changing role of engineers, and the profound implications for business models as the industry pivots from providing services to delivering concrete, measurable solutions.
As software begins targeting the $50 trillion global labor market, how must vendors fundamentally change their product strategy and pricing models to convert traditional labor costs into recurring software revenue? Can you share a specific example of this shift in action?
It’s a complete paradigm shift. For decades, software vendors were fighting for a piece of the corporate IT budget, which is a massive market at $1.3 trillion, but still a defined slice of the pie. Now, with AI, the thinking is so much bigger. We’re looking at the entire $50 trillion global labor market and asking, “Which tasks can be converted from human labor costs into a software subscription?” The product strategy must pivot from creating tools that make employees more efficient to building “digital workers” that can replace entire functions. Imagine a law firm. Instead of paying paralegals for contract review, they now subscribe to an AI platform that does it. The vendor isn’t selling a better word processor; they’re selling the output of a paralegal, effectively turning a salary line item into recurring software revenue.
When a CRM or ERP system evolves from a simple record-keeper into an active “digital worker” that makes decisions, what are the biggest operational challenges a company faces? Please walk us through the steps to manage that transition successfully.
The biggest challenge is trust, followed closely by process integration. For years, an ERP was a trusted, if passive, digital filing cabinet; its value was in its historical accuracy. When that system starts making active decisions, like forecasting cash flow or identifying high-value leads on its own, it introduces a massive operational hurdle. People are naturally resistant to ceding control. The first step to manage this is starting small with a pilot program to prove its reliability. Second, you absolutely must tackle data hygiene. AI is only as good as the data it’s fed, so cleaning and mapping old data is critical for the system to make sound judgments. Finally, you have to redesign the surrounding human workflows. It’s not about plugging in an AI and walking away; it’s about redefining roles so that employees are managing the outcomes the AI produces, not just doing the manual work it replaced.
With AI assistants enabling smaller engineering teams to deliver more, how does the role of the modern software engineer change? What specific skills should they prioritize to become architects of intelligent systems rather than just coders?
The role is being elevated dramatically. We’re moving past the era where product growth meant linear headcount growth in engineering. AI coding assistants can now handle routine code generation, bug detection, and even modernizing legacy systems, which used to consume enormous amounts of an engineer’s time. This frees them from the grunt work. Consequently, the most valuable engineers are no longer the ones who can write the most lines of code the fastest. They are the architects, the strategic thinkers. They need to prioritize systems-level thinking, understanding how to integrate different AI models and data sources to solve a core business problem. Domain expertise also becomes paramount; an engineer who deeply understands the intricacies of, say, insurance claims processing will be far more effective at building an intelligent automation system for it than a pure coder. The key question for companies has shifted from “Do we have enough engineers?” to “What should we build next?”
The industry is shifting from Software-as-a-Service to “Software-as-a-Solution.” What are the biggest hurdles in pricing a product based on a measurable outcome, like “per sales lead generated,” and how does this change the dynamic between the vendor and the customer?
The biggest hurdle is managing variability and risk. Traditional SaaS offered predictable, per-seat subscription revenue, which investors love. But when you price “per ticket resolved” or “per meeting booked,” you introduce new variables. Your own computing costs can fluctuate, and the customer’s usage patterns are less predictable. Customers are also wary; they are hesitant to pay fixed fees for AI capabilities that haven’t proven their worth yet. This fundamentally changes the dynamic into a true partnership. The vendor is no longer just a tool provider; their success is directly tied to the customer’s success. This forces a much deeper level of collaboration and alignment on business goals, but it also means the vendor takes on more performance risk.
AI software often enters an organization by solving one small, high-impact task for a single department. Could you describe the typical journey of such a tool, from initial adoption to becoming an indispensable, company-wide platform, highlighting the key inflection points?
It’s a classic “land and expand” strategy, but supercharged by AI’s ability to learn and improve. It almost never starts with a massive, company-wide installation. Instead, it enters through a side door. A department head might approve a small budget for an AI voice agent to handle after-hours customer calls, feeding the results into the main CRM. This delivers a quick, visible win. The first inflection point is when that tool starts to demonstrate its reliability and ROI. As trust grows, its role expands. That voice agent might then be tasked with handling routine daytime calls, sorting tickets, and eventually evolving into a full-blown case management system. The second, and most critical, inflection point is when other departments see the success and want a piece of the action. What began as an invoice assistant in accounting can grow into a fully automated accounts-payable platform for the entire organization. It becomes a core platform not by force, but by proving its value one tangible outcome at a time.
Given that in-house innovation with AI is becoming cheaper and faster, what are the primary risks a large software company now faces when acquiring an AI startup? Under what specific circumstances does pursuing an acquisition still make strategic sense?
The primary risks are cost and talent retention. AI startups command huge premiums, and the real value isn’t in the code, which can be replicated, but in the highly specialized engineering talent. The moment the acquisition closes, the countdown begins on those top engineers leaving after their contracts are up, which drastically reduces the long-term value of the deal. Furthermore, AI can’t just be bolted on; it has to be deeply woven into a company’s core products and workflows, which is a difficult integration challenge. With AI tools making internal R&D so much faster and cheaper, building is often more attractive than buying. An acquisition still makes strategic sense only in very specific cases—for instance, when a startup possesses a truly unique, defensible dataset or a breakthrough proprietary model that would take years to replicate. Otherwise, the focus is shifting back to innovation as the primary engine of growth, not acquisition.
What is your forecast for the software industry over the next five years?
Over the next five years, I forecast a fundamental re-platforming of the industry centered on outcomes rather than features. We will see the most successful companies be those that effectively transform labor costs into software revenue, positioning their products as “digital workers.” This will force a massive shift in pricing and sales models, moving away from per-seat licenses to usage- and value-based pricing, making the vendor-customer relationship much more of a direct partnership. We’ll also witness a renaissance in internal R&D, as AI tools empower smaller, more agile teams to out-innovate larger, slower competitors, which will in turn make the M&A landscape far more selective and strategic. Ultimately, the very definition of software will have changed; it won’t be a tool you use, but a solution that delivers a quantifiable business result.
