Services-as-software: 5 ways SaaS firms should adapt

Key Takeaways

• VC Orlando Bravo declares the 'SaaS apocalypse' over, calling AI an enormous tailwind for software companies
• Enterprises can't deploy AI at scale yet, keeping traditional SaaS relevant while a hybrid model emerges
• Software and services firms are converging on outcome-based pricing, creating the 'services-as-software' model

Venture capitalist Orlando Bravo has declared the so-called SaaS apocalypse over. In a recent CNBC interview, the Thoma Bravo founder argued that AI isn't killing software companies. It's accelerating them. "AI is an enormous tailwind for software companies," Bravo said. "Software companies can move to a completely new level of business automation by automating some parts of human judgment."

The numbers back him up. Salesforce, the company most synonymous with SaaS, reported $11.1 billion in quarterly revenue in May 2026, a 13% year-over-year increase. The company then acquired customer service software firm Fin (formerly Intercom) for $3.6 billion. Even IBM, which just posted a brutal quarter overall, saw 5% growth on its software side.

But here's the twist: SaaS isn't staying the same. It's morphing into something new. Analysts Saurabh Gupta and Phil Fersht of consulting firm HFS call it "services-as-software." The distinction matters for anyone building or running a software business.

Why enterprises aren't abandoning SaaS yet

The AI-replaces-software narrative has a problem: enterprises can't actually deploy AI at scale. "Until organizations resolve their technology, data, process, and talent debt, AI will remain trapped in pilots and proofs of concept rather than fundamentally changing how businesses operate," Gupta and Fersht write.

This is the gap between Wall Street predictions and Main Street reality. Tech investors declared SaaS dead while actual businesses are still figuring out where AI fits. Most will stick with their existing SaaS tools, including platforms like HubSpot, Pipedrive, and Zoho CRM, for the foreseeable future. The infrastructure debt is too deep to clear overnight.

What is services-as-software?

The shift isn't from SaaS to something else. It's a convergence. Services firms are becoming software businesses. Software companies are moving deeper into implementation and business transformation. Both are landing on outcome-based economic models.

IBM illustrates this. The company has become, in Gupta and Fersht's words, "a software and AI business that happened to own a consulting arm, rather than a consulting business trying to sell AI." The distinction is subtle but significant. The software is the core; the services wrap around it to drive adoption.

The winners in this model will be firms that "combine AI with deep client relationships, transformation expertise and privileged access to enterprise systems." Pure-play software vendors without implementation muscle may struggle. Pure-play consultancies without software IP may also find themselves squeezed.

5 ways to adapt to services-as-software

Gupta and Fersht offer specific guidance for software companies navigating this transition:

1. Treat enterprise debt as an up-front business issue. Measure, prioritize, and fund technology, data, process, and talent debt "with the same discipline you apply to capital investments."
2. Focus on business outcomes, not pilots. "If an AI initiative cannot demonstrate meaningful commercial impact within 90 days, question whether it deserves further investment. Every failed pilot delays the transformation you're actually trying to achieve."
3. Buy outcomes instead of effort. Think business value, not licenses, tokens, or full-time equivalents. "If a supplier cannot explain how they improve your P&L, they are selling technology rather than transformation."
4. Align your AI and services partners. "If they are working independently, you will pay for the disconnect."
5. If you're with an AI-native firm, earn trust before expecting scale. New entrants need to prove value in contained engagements before enterprises will commit to broader deployments.

How pricing models are changing

The services-as-software model has direct implications for how software gets priced. Traditional SaaS relies on seat-based or usage-based pricing. The emerging model ties revenue to business outcomes.

This isn't entirely new. Enterprise software contracts have always included implementation fees and success metrics. What's different is the expectation that software vendors take more responsibility for outcomes, not just uptime. A CRM that increases sales conversion rates by a measurable percentage is worth more than one that simply logs contacts.

The shift rewards vendors who can prove impact. It punishes those who ship features without tracking whether anyone uses them.

What this means for SaaS founders

If you're building a SaaS product today, the services-as-software model suggests a few strategic moves. First, think about your implementation story from day one. Can customers deploy your product and see results in 90 days or less? If not, you'll need partners or internal capacity to bridge that gap.

Second, build measurement into your product. If you can't show customers the business impact you're delivering, someone else will. The vendors who can demonstrate P&L improvement will win the outcome-based contracts.

Third, consider where AI fits. Not as a feature checkbox, but as a way to automate parts of human judgment that currently require services. That's the tailwind Bravo is talking about. Software that can do more of the work, not just organize it, is where value is heading.

Frequently Asked Questions

Is SaaS really dying?

No. The 'SaaS apocalypse' narrative was overblown. Salesforce posted 13% revenue growth in Q1 2026. What's changing is how SaaS is delivered, with more emphasis on implementation services and outcome-based pricing.

What is services-as-software?

A hybrid model where software vendors provide implementation, transformation, and ongoing services bundled with their products. Revenue is tied to business outcomes rather than just seats or usage.

Why can't enterprises deploy AI at scale yet?

Most organizations have significant technology, data, process, and talent debt that prevents AI from moving beyond pilots and proofs of concept. Until that debt is addressed, traditional SaaS remains necessary.

How should SaaS pricing change for outcome-based models?

Vendors need to tie pricing to measurable business impact, not just features or usage. This requires building analytics into products that track customer outcomes like revenue growth or cost reduction.

Source: Latest news