Vijay Raina is a seasoned veteran in the SaaS landscape, known for navigating the complex intersection of software architecture and revenue strategy. As enterprises grapple with the massive platform shift triggered by agentic AI—a transition representing over $3 trillion in potential revenue—Raina has become a critical voice for companies trying to avoid the existential threat posed by changing market dynamics. In this conversation, we explore how the sudden market volatility of early 2026 served as a wake-up call for SaaS providers to rethink their moats, their pricing, and their very foundations. We delve into the rising pressure from AI-native startups that scale at unprecedented speeds, the necessity of building “pivot-ready” infrastructure, and the psychological shift required when machines, rather than humans, begin to act as the primary buyers of software.
The software market experienced a massive shock earlier this year with a $2 trillion loss in value. What was the underlying cause of that panic, and why did it strike such a chord with enterprise providers?
The tremor we felt in early 2026 was a moment of realization that the traditional SaaS moat—the complex, expensive features we spent years building—is being eroded by AI-powered coding tools that allow rivals to copy functionality in a fraction of the time. There was a palpable fear among investors that agentic AI would fundamentally undermine the business models we’ve relied on for a decade. We saw US software stocks lose about $2 trillion in value because the market began to view application software as facing an existential threat rather than a growth opportunity. It wasn’t just a random dip; it was a reaction to the fact that 66% of North American and European buyers are now shifting their budgets, planning to spend more on AI and agents while aggressively consolidating vendors and cutting costs everywhere else. This creates a “seat compression” effect where the old per-seat subscription model starts to feel like a liability rather than an asset.
AI-native companies seem to be moving at a speed that traditional SaaS firms struggle to match. How significant is the performance gap between these new players and the incumbents?
The data we are seeing is honestly staggering and should serve as a loud siren for any legacy provider. AI startups founded within the last three years are reaching the $5 million annualized revenue milestone in a median of just 9 months, which is a lightning-fast trajectory compared to the 37 months it took the top 100 SaaS companies of the previous era to reach that same level. Even at the $1 million mark, the top 100 AI companies are hitting that target in a median of 11.5 months, roughly 4 months faster than the most elite traditional SaaS firms. These AI-native players have a leaner internal economic structure and often use AI to boost their own productivity, meaning they don’t have to increase headcount at the same rate they increase revenue. They are built from the ground up to be “pivot-ready,” allowing them to experiment with monetization from day one without the fear of cannibalizing an existing, multi-million dollar revenue stream.
You’ve mentioned that SaaS providers need to become “pivot-ready.” What does that concept actually look like in practice for a company that has spent years on a standard subscription model?
Being pivot-ready is less about a single pricing change and more about creating the organizational flexibility to test, adapt, and implement new models like usage-based or outcome-based pricing on the fly. It is a vital survival trait in an era where compute costs are no longer a background issue but a volatile variable that fluctuates based on token usage and third-party LLM pricing. We’ve found that 67% of hypergrowth companies have changed their pricing three or more times in the last two years, compared to only 22% for low-growth companies. To achieve this, a company has to treat its billing and monetization as core infrastructure rather than just an administrative back-office task. It requires building telemetry that can meter usage at the event level, decoupled from the actual product code, so that a pricing experiment doesn’t require a six-month engineering sprint to launch.
As companies integrate agentic AI, they often struggle to show the actual value to the customer. How should providers restructure their “taxonomy of value” to ensure they aren’t just offering “table stakes” features?
This is a major hurdle because many AI features we see today are merely convenient time-savers for individuals—like summarizing a meeting—but they don’t move the needle on enterprise-level ROI. To win customers all over again, providers need to create a hierarchy that distinguishes these basic “table stakes” from high-value agentic capabilities that actually drive down operating costs or generate revenue. We often see a “value attribution” gap; in our surveys, the top reason companies shy away from outcome-based pricing is that they simply don’t measure the outcomes or can’t prove the platform was responsible for them. By building a clear path that connects platform activity to financial or operational results, a provider can justify a shift away from seat-based models and toward a model where they share in the success they create for the client.
Compute costs and token pricing are notoriously unpredictable. How does a SaaS company protect its margins without scaring off customers with complex, fluctuating invoices?
The “token cost curve” is a bit of a roller coaster right now, with unpredictable twists and turns that make linking revenue directly to tokens a very risky move—it’s like tying your business’s health to a fairground ride. Instead of using token counts as a primary metric, we recommend using them as secondary indicators while moving toward a hybrid pricing model. We’ve found that 49% of high-growth companies and 57% of hypergrowth firms have already adopted these hybrid models, which combine the predictability of a base subscription with a usage or outcome-based component. This provides the “AI wallet” with the budget certainty that CFOs crave while allowing the provider to capture the upside of increased AI engagement. It prevents the provider from being buried by compute costs if a customer suddenly ramps up their agentic activity, but keeps the invoice understandable enough to maintain trust and avoid nonpayment risks.
We are beginning to see “headless” sign-ups where agents, not humans, are interacting with software. How does this shift change the fundamental psychology of pricing?
When you remove the human from the purchasing process, traditional psychological tactics like “price anchoring” or the “good-better-best” tiering lose their effectiveness because an AI agent doesn’t have a human brain that defaults to the middle option. These agents can evaluate thousands of combinations of vendors and tiers with ruthless efficiency to find the exact configuration that yields the most value for the lowest cost. We are already seeing this with platforms like PostHog, which reported more than 200 headless sign-ups in 2026, many of which were entirely agent-driven through command-line interfaces. SaaS players now have to think about how to charge for API access and fair use in a way that satisfies a machine buyer that might rebalance its subscription and usage options every single month. It erodes the revenue predictability we’ve enjoyed for years, forcing us to compete on the actual efficiency and utility of the data and orchestration we provide.
What is your forecast for the SaaS industry as we move through the remainder of 2026 and into 2027?
My forecast is that we will see a Great Bifurcation between incumbents that successfully transition to hybrid, value-aligned pricing and those that remain tethered to the dying seat-based model. We will likely see a surge in “transactional platforms” where SaaS providers start to coordinate multiple ecosystem partners, providing embedded finance and payment capabilities as part of their core offering to capture value beyond the software itself. By 2027, the concept of a “user seat” will likely feel as antiquated as “buying a CD-ROM” did in the early 2010s, as agentic managed services become the primary way enterprises consume technology. The companies that thrive will be those that viewed the $2 trillion market dip of early 2026 not as a temporary setback, but as the starting gun for a total reinvention of how software creates and captures financial value.
