Vista's Robert Smith on AI Revolution in Enterprise Software

In Brief:

Enterprise software stands at the threshold of its most profound transformation since the internet's dawn, with artificial intelligence poised to fundamentally reshape profit margins, workforce composition, and competitive dynamics across the industry. Robert F. Smith, founder, chairman, and CEO of $100 billion Vista Equity Partners, argues that companies lacking "sovereignty and dominion" over their proprietary workflows and data sets face existential threats in this new era, while incumbents positioned correctly could see operating margins surge from 25% to 40% and beyond. In a wide-ranging conversation on the Alt Goes Mainstream podcast, Smith—who built Vista into the world's fourth or fifth largest enterprise software company by revenue after pioneering the sector as a Goldman Sachs tech banker in the 1990s—draws on his chemical engineering background and three decades of pattern recognition to explain why AI won't eat software, but rather will feed software that eats services, creating what he estimates as 8-10 billion software agents displacing traditional knowledge work.

Big Picture Drivers:

• Engineering mindset foundations: Smith's chemical engineering background at Cornell taught him to think about systemic process improvements rather than episodic interventions, seeing early how compute technology could increase plant productivity by 26% through tighter control paradigms

• Enterprise software fundamentals: The sector offers mission-critical business applications with 95% gross margins, negative working capital, no inventory, and high customer retention rates that create stable base-loaded returns with convexity on the upside

• Technology transformation waves: From on-premise to hosted to cloud infrastructure over 20 years, each wave created new opportunities to capture economic rent by reducing customer costs while adding functionality

• AI's infrastructure parallel: Current AI adoption mirrors the internet's evolution—hardware vendors like Nvidia capture early gains, hyperscalers build infrastructure, but enterprise software vendors with proprietary workflows and data will ultimately capture the majority of economic value

Key Themes:

• Data sovereignty as competitive moat: Success in the AI era fundamentally depends on maintaining control over proprietary workflows and data sets, with less than 1% of enterprise data currently residing in public LLMs, creating lasting advantages for companies that protect their information assets

• Operational transformation through AI: Generative AI can reduce product development costs from 30% to 20% of revenue, deflect 60-90% of customer support calls with higher satisfaction scores, and enable salespeople to spend 40-70% more time actually selling versus administrative tasks

• Contract quality as economic foundation: Enterprise software companies are only as valuable as the quality of their signed contracts, yet 99% of software executives don't list contracting among their top five priorities, creating systematic value leakage

• Agentic AI as workforce multiplier: The shift from 750 million software users to 8-10 billion software agents represents Vista's strategic pivot from the "user business" to the "agent resources business," with agents functioning as 24/7 workers when properly built, tuned, and enabled

Key Insights:

• The Rule of 40 becomes the Rule of 70: Vista coined the "Rule of 40" metric combining EBITDA margin and growth rate, but generative AI enables companies to achieve Rule of 70 performance by simultaneously reducing development and support costs while increasing sales productivity without selling additional products.

• Services markets face displacement: AI-enabled software won't replace software itself but will feed software applications that displace traditional services markets, as agentic workers perform tasks previously requiring human knowledge workers at vastly greater scale and speed.

• Enterprise AI requires precision impossible for consumers: While consumer AI can tolerate 5% error rates (booking the wrong restaurant), enterprise applications demand near-perfect precision in workflows like payroll and financial transfers, making deployment significantly more complex than personal productivity use cases.

• Incumbents have natural AI advantages if managed correctly: Companies that have maintained sovereignty over workflows and data sets possess structural advantages in the AI era, though many CEOs compromised this position during the ARR growth race by leaking intellectual property through poorly structured contracts.

• Vista's "bring models to data" architecture: Rather than sending proprietary data to external LLMs where it gets absorbed and benefits others, Vista brings AI models into their controlled environments, ensuring companies retain exclusive utility and benefit from their unique data assets.

• Contract variance undermines economics systematically: Average salespeople spend 95% of their time convincing their own company to accept customer-friendly contracts rather than persuading customers to accept company-favorable terms, creating systematic economic degradation across the portfolio without intentional process controls.

Memorable Quotes:

• "If you don't have sovereignty and dominion over workflows and data sets, you don't have a right to exist as an enterprise software company. If you do, then you have a right to dominate." - Robert F. Smith, explaining the existential stakes of data control in the AI era

• "AI is not going to eat software. AI is going to feed software that's going to eat services." - Robert F. Smith, articulating his thesis on how AI transforms market dynamics and addressable opportunities

• "Enterprise software companies' economic position is defined by their contracts and you're only as good as the quality of the contracts that you have signed." - Robert F. Smith, articulating why contract quality represents the foundation of enterprise software value

• "Chemical engineers are designed to create elegant solutions to complex problems. And if you take that paradigm of thought and infuse it into investing, that's really what Vista is." - Robert F. Smith, connecting his engineering background to Vista's systematic approach to value creation

• "I'm shocked that CEOs haven't done a good job preserving sovereignty and dominion. They were on an ARR race, chasing growth in a way that they've leaked intellectual property and ownership of the very thing that would enable them to be more successful in an agentic world." - Robert F. Smith, warning about strategic mistakes made during the growth-at-all-costs era

The Wrap:

Smith's framework for understanding AI's impact on enterprise software reveals that this transformation differs fundamentally from previous technology waves in its speed and completeness. While the shift from on-premise to cloud took 16 years and still covers only 50% of enterprise software, AI's operational improvements manifest immediately in companies that implement it correctly. The existential divide between winners and losers will be determined by something most executives don't even prioritize: whether they've maintained contractual sovereignty over their customer workflows and data. For Vista's institutional and wealth channel investors, this creates both urgency—to identify and fix data leakage in existing portfolio companies—and opportunity, as properly positioned software companies could see margins and growth rates that rewrite traditional private equity return expectations. The ultimate irony is that a 24-year-old commanding six to eight agents could destroy large software companies that failed to protect their data moats, while Vista's systematic approach to contracts, processes, and data architecture positions its 90 portfolio companies employing 100,000 people to capture disproportionate value in the agentic era.