SaaS Disruption Through AI-Driven Models

The Death of SaaS (as We Know It) Satya Nadella recently shared a fascinating perspective: AI is poised to replace traditional application layers, embedding business logic directly at the database level. This marks a profound shift: one that could redefine the very foundation of SaaS. Imagine a future where AI doesn’t just power apps but replaces them. Business logic, instead of flowing through multiple layers of UI, middleware, and APIs, is orchestrated directly with the database. This means the end of bloated, layered software and the beginning of lean, AI-native architectures. The ripple effects are massive. SaaS as a subscription model may lose relevance as modular AI-driven workflows dominate. Interfaces will transform, shifting away from dashboards and fixed workflows to adaptive, real-time experiences—think voice commands, conversational AI, or neural interfaces. Even the app store economy may collapse under the weight of this new paradigm, replaced by marketplaces for AI-driven workflows instead of apps. This could imply the extinction for the SaaS we know today. For developers, businesses, and consumers, this shift will reshape how software is built, sold, and used. The question isn’t if SaaS is dying; it’s what comes next. What do you think? Is this the end of SaaS, or the beginning of something even more disruptive?

If I were running a legacy SaaS company today, I wouldn’t be sleeping much. For legacy SaaS startups, pivoting to an AI-native company is an existential challenge, testing the core of the Innovator's Dilemma. To their credit and courage, most SaaS CEOs are taking action, yet far too incremental, taking an "AI 1.0" approach by adding a copilot to their existing product. Real transformation lies in "AI 2.0"—reimagining the fundamental user interaction from the ground up. Why the alarm bells are ringing? * AI 1.0 ≠ transformation. Most SaaS incumbents bolt on a “copilot”. Nice demo, small impact. * AI 2.0 re-imagines the interface and workflow. Think GitHub Copilot vs Cursor: autocomplete add-on vs. full-stack code co-author that rewrites files, reasons across repos, and adapts to any model — developers feel the difference instantly. *The system-of-record moat is eroding. SaaS data model-based moat that created stickiness for the last two decades—is being replaced by conversational, intent and agentic based systems. Example:  CRM goes from a database to completing RFPs and follow-up emails. Why Legacy SaaS default to AI 1.0? - SaaS CEOs overestimate stickiness of the current UX and data model.  Customers will migrate. - Underestimate CIO/CTO AI mandates (new AI budgets are cannibalizing legacy line items). - Culture favors incremental roadmaps over zero-to-one bets. How Legacy SaaS can build for AI 2.0? 1. Redesign the interface. Start with the work-to-be-done, not the existing SaaS interface. 2. Build an orchestration layer for agentic workflows, tool calling, and human in the loop. Your current middleware gives a head start; extend it. 3. Staff for 0→1. Put founder-type product & engineering leaders, perhaps in an autonomous pod. Protect them from quarterly roadmap gravity. 4. Incentivize Customer Migration.  Ensure incentives of GTM teams are aligned to upgrading and moving existing customers over to the new platform.  Leadership test Ultimately, this is a test of leadership.  The SaaS CEOs and Founders who win will be those with the conviction to build for a new reality, even if it means disrupting their own successful products.

A new Fortune piece out with The Wharton School's Stefano Puntoni asks a simple question: is AI just improving #enterprise_software — or is it starting to \dismantle the business model that made SaaS so profitable in the first place? My view: the term “#SaaSpocalypse” may be exaggerated, but the underlying pressures are real. Based on two recent roundtables with senior business leaders in New York and San Francisco, we argue that three structural shifts are now converging. First, AI is exposing long-hidden market #vulnerability. For years, many enterprise software firms have enjoyed margins supported not only by product quality, but by #switching_costs, #customer_inertia, and the sheer #pain of migration. That is a powerful model — but also a fragile one. When customers remain because they feel trapped rather than delighted, #disruption becomes much easier to imagine. Second, #barriers_to_entry are falling fast. Building enterprise-grade software used to require major capital, large engineering teams, and long development cycles. AI coding tools are beginning to change that equation. This does not mean incumbents disappear overnight. But it does mean more entrants, more #experimentation, more credible #alternatives, and more pressure on the economics the sector has long taken for granted. Third, and perhaps most importantly, AI is changing what #customers actually #value. SaaS became powerful by #standardising_workflows across firms and sectors. But AI allows workflows to be #rethought from the ground up. That may shift advantage away from generic horizontal tools and toward context-rich, sector-specific #intelligence. In other words, deep #vertical_understanding may become more valuable than broad process standardisation alone. There is also a wider #ecosystem_story here. As copilots and agents begin to configure, run, and reshape workflows, the old division of labour starts to blur: software vendors, systems integrators, consultants, hyperscalers, and model providers are all moving into one another’s territory. The battle shifts to new control points — #orchestration, privileged #data_access, and distribution into day-to-day work. So no, I do not think enterprise software is disappearing. But I do think many of the assumptions that made it such an attractive and profitable sector are now under pressure. Link to a paywall-free version in comments. PS: More on ventures in this space am involved in to follow soon (shoutout, Evolver and Mario Schlener, Luis Vargas, PhD, Bas Kamphuis, Nicola Morini Bianzino); ditto for thoughts on how major SaaS incumbents should respond.

If 2023-2024 was about "chatting" with your data, late 2024 and 2025 are about acting on it. The biggest players in SaaS- Salesforce, Microsoft, and Google have all pivoted their strategy toward autonomous agents that don’t just summarize meetings but execute workflows. Here is the breakdown of the current landscape for enterprise leaders: 1️⃣ The "Big Three" Are All-In on Autonomy Salesforce: With Agentforce, they are betting the farm on agents that can autonomously resolve customer service tickets and qualify sales leads, moving beyond the "human-in-the-loop" bottleneck. Microsoft: Copilot Studio has evolved to let you build autonomous agents that work in the background, triggering supply chain updates or IT tickets without a user prompt. Google: The Vertex AI Agent Builder and Gemini 2.5/3 updates focus on "Agent2Agent" protocols, allowing different AI agents to talk to each other to solve complex problems without human intervention. 2️⃣ The "Pilot Purgatory" Reality Check: Despite the hype, the data tells a mixed story. While ~78% of organizations have adopted AI in some form, a recent study suggests only a small percentage are seeing significant revenue growth from it. The Problem: Most enterprises are stuck in "Pilot Purgatory." It’s easy to build a demo; it’s incredibly hard to govern an agent that can delete records or send emails on its own. The Barrier: Data silos. An agent is only as good as the data it can access. If your data is messy, your agent isn't a "coworker", it's a liability. 3️⃣ The Death of the "Per-Seat" Model? This is the silent disruption that is happening in the background. If an AI agent does the work of three SDRs, why should Salesforce or Microsoft charge you for just one "seat"? We are seeing a shift toward consumption-based or outcome-based pricing. You pay for the work done (e.g., tickets resolved, conversations handled), not the number of humans logging in. 💡 The Takeaway for Enterprise Leaders: Stop looking for "AI features" in your SaaS contracts. Start looking for Agentic Frameworks. Don't ask: "Can this software summarize a PDF?" Ask: "Can this software autonomously execute a 5-step workflow with my governance rails in place?" The tools are ready. Is your data layer there? #AI #SaaS #AgenticAI #DigitalTransformation #TechTrends2025

The market has suddenly realized that AI is breaking SaaS economics, not just speeding up automation. With the launch of Claude Code and OpenClaw, a flood of Substacks, technical blogs, and macro research reframed the conversation around AI’s economic impact on software. Real production examples showed agents booking meetings, writing code, generating designs, handling support, and running workflows end-to-end. Investors connected the dots: fewer humans mean fewer software seats, fewer enterprise licenses, and compressed margins. That’s why ServiceNow, Salesforce, Adobe, HubSpot, and Workday sold off hard. Their models assume humans using software. Agents remove humans from the loop. That’s existential, not cyclical. At the same time, blogs reframed Indian IT revenue as: people × billing × project duration. AI attacks all three. Fewer engineers, pressure on pricing, shorter projects. Investors modeled the same clients and contracts, but a much lower long-term revenue slope. So Infosys, Tata Consultancy Services, HCLTech, LTIMindtree, and Coforge repriced together. India exports human effort. AI replaces human effort. The market discounted earnings durability, not next quarter EPS. Another theme took hold: profits concentrate with foundation models and compute, not with integrators or SaaS middlemen. That’s why NVIDIA held up relatively better and Microsoft stayed more resilient, while services and application software de-rated. The conclusion was simple: AI is centralized, not distributed. Bad for everyone in the middle. Citrini-style macro essays added a darker layer. If AI destroys white-collar jobs faster than it creates new ones, consumer demand weakens, enterprise budgets shrink, and tech spending slows. AI creates a self-reinforcing deflationary loop where automation reduces employment, weakens demand, forces more automation, and concentrates profits upstream, while markets are still valuing companies as if this is a normal business cycle. Investors like Michael Burry say that this isn’t a normal tech cycle. It’s a business-model reset. Markets aren’t reacting to earnings misses, they’re repricing durability. SaaS is being valued as if seats disappear. Indian IT is being valued as if labor stops scaling. Capital is flowing upstream to compute and models. Everything in the middle is being questioned. The mistake investors will make is treating this like another automation wave. It isn’t. This is workflow deletion. And once that lens sets in, multiples don’t bounce back easily.

We’re watching the rapid transformation - and possible end - of SaaS as we know it. Microsoft CEO Satya Nadella recently pointed out that traditional SaaS is disappearing, and I strongly agree. But I see the timeline accelerating even faster: Phase 1 (Right now): AI as Support AI enhancements like Copilot, Gamma, and Harvey are currently complementing existing SaaS platforms, making them seem more efficient and attractive. Providers feel secure, viewing AI as a feature rather than a threat. Phase 2 (Within 6-12 months): AI Takes Over Operations AI agents will quickly transition from assistants to autonomous operators. Instead of manually using tools like Tableau or Meta’s ad platform, we’ll simply instruct agents to perform analyses or optimize ads directly. The expertise traditionally embedded in SaaS interfaces becomes easily accessible through agents. Phase 3 (Within 1-2 years): Software Becomes Invisible AI agents begin interacting directly via APIs, eliminating the need for human-oriented interfaces like dashboards and menus entirely. This strips away the core value SaaS once provided—human usability. This isn’t standard disruption; it’s a fundamental shift away from human-operated software to agent-operated software. At the same time, the rise of AI-driven coding tools makes custom internal software development dramatically easier and cheaper. Companies no longer need to rely on costly SaaS subscriptions—they can quickly create tailored internal applications that perfectly fit their needs. The winners in this new era won’t simply be those who integrate AI the quickest. Instead, they’ll be companies providing open, agent-friendly APIs, becoming the trusted providers of actionable data and execution within their fields. The real question is whether giants of all industries will swiftly adapt or risk becoming obsolete, much like tech giants of the past. We’re entering an extraordinary period of opportunity for agile startups ready to embrace this change.

The past few weeks have provided a sobering reality check for the software industry. The recent, brutal drops in SaaS and security software valuations are not just panic reactions to AI progress. They reflect a fundamental shift in how business value is created in the age of AI. Claude Cowork and OpenClaw projects show that the future lies not in application-specific agents but in the ability to coordinate agentic workflows centrally. It is not good news for SaaS companies that add agents to their offerings. Microsoft CEO Satya Nadella said more than a year ago that the traditional SaaS model is becoming obsolete, and software companies must pivot to AI agents or risk fading into irrelevance. But even pivoting to agents may not be enough. In my recent conversations with enterprise leaders, the sentiment seems nearly unanimous. Many CIOs and CFOs have explicitly told me they plan to rip out up to hundreds of SaaS applications this year. The era of buying a specialized point solution for every minor business problem is over. Leaders are moving toward ruthless consolidation and are clearly focusing on building a competitive, company-controlled data layer rather than outsourcing their data management to SaaS applications. The architectural shift underneath this consolidation is a profound technical migration. For decades, businesses operated on an application-centric model, where data was fragmented and trapped behind dozens of different user interfaces and proprietary business logics with complex integrations. We are now moving rapidly toward a data-centric architecture. In this new paradigm, data sits at the secure core of the business. Autonomous AI agents interact with that data directly to drive outcomes, often bypassing the need for a traditional software GUI entirely. When the interface matters less, the per-user subscription model tied to it loses its justification. This shakeout is going to be massive. It will be a painful transition for many companies that built incredible products based on the old rules of user-based licensing. But there's a silver lining for businesses that lead rather than follow this transformation. Increased data sovereignty: As enterprises shed redundant applications and centralize their architecture, they are reclaiming ownership of their information. You will no longer be forced to rent your workflows and scatter your data across fifty different third-party vendors. A new competitive edge: The corporate battleground is shifting. Your competitive advantage will no longer be defined by industry-standard applications, but by the quality and structure of your proprietary data and by how effectively you deploy your agent swarms to act on it. The software landscape is fundamentally transforming. It is a difficult pivot, but the businesses that lean into this shift will emerge leaner, smarter, and entirely in control of their own destiny. Would you agree?

𝐒𝐚𝐚𝐒 𝐚𝐬 𝐰𝐞 𝐤𝐧𝐨𝐰 𝐢𝐭 𝐢𝐬 𝐝𝐢𝐬𝐚𝐩𝐩𝐞𝐚𝐫𝐢𝐧𝐠 In his conversation with Sam Altman and Brad Gerstner, Satya Nadella explained that traditional SaaS apps — built around static logic layers and human users — are being replaced by AI agents that perform the same workflows autonomously. Instead of humans clicking through a CRM, ERP, or project management tool, agents will sit on top of the data, understand the context, and take action. Nadella described it as a structural shift: The old SaaS stack (data + logic + UI) was tightly coupled. The new stack separates the AI logic layer from the interface, turning agents into the new users. Usage patterns flip — from “per seat” pricing to “per agent” consumption. He added that in Microsoft’s products — from GitHub Copilot to Microsoft 365 Copilot — usage and data creation have surged. The more AI is integrated, the more data is generated, which in turn powers better grounding for future models. In this new model, agents become the interface, and data becomes the product. SaaS isn’t ending — it’s evolving into something more autonomous, contextual, and continuous.

I’ve been building and investing in software companies for over two decades—from early SaaS startups to multi-exit outcomes. Every cycle brings hype, but what’s happening now with agentic AI feels like a deeper inflection point. It’s not just another tech wave. It’s a redefinition of how work gets done. This isn’t about better AI assistants or copilots. It’s about agents that execute end-to-end tasks without human handoffs. The implications go far beyond productivity. A startup I met recently replaced several marketing ops contractors with an agent that can ingest performance data, generate copy, spin up landing pages, A/B test variants, and tune campaigns. All from a single prompt like, “Re-engage churned users this month.” No one logged into a dashboard. No one ran a spreadsheet. No Zapier. It just worked. Another founder I know is using agents to manage QA—finding bugs, writing test cases, suggesting fixes, and even shipping patches. That’s not augmentation. That’s replacement. Naturally, this means labor cost savings. Especially in repeatable, rules-based roles, agents do the job faster, cheaper, and (often) more accurately. For startups and lean teams, it’s a superpower. For enterprises, it’s a massive margin lever. But it also raises hard questions: - What happens to jobs when the “user” is no longer a human? - What do org charts look like when an AI agent can run an entire function? - How do we preserve human oversight and accountability in a system increasingly run by machines? - As someone who's hired, mentored, and backed hundreds of people in this industry, I don’t ask these questions lightly. There’s a real risk of organizations chasing cost reduction without thinking through long-term cultural, ethical, and societal implications. For tech leaders—CIOs, CTOs, CDOs, CPOs, CISOs—this is the next wave of disruption. If your architecture, org design, and security model aren’t built for autonomous agents, you're behind. For SaaS founders, it’s time to think about product design when the “customer” is software, not just people. Some companies will adapt. Some will get unbundled. Others may simply become APIs called by agents built elsewhere. There’s opportunity here—but also responsibility. Let’s not just ask what we can replace, but also what we should. How are you seeing this show up in your work or teams as a tech exec / operator or startup founder? #AgenticAI #FutureOfWork #SaaS #EnterpriseSoftware #CTO #CIO #CISO #Startups #AITransformation #Founders #GenAI

Are Software Applications Headed Toward Extinction? For 40 years, enterprise computing has followed one dominant model: - Build software applications - Force workflows into the application - Train humans to adapt to the system From ERP to CRM to modern SaaS — the paradigm never really changed. But Generative AI and AI agents may fundamentally break this model. A question worth asking: Why do we need software applications at all? Today, if I want to: - analyze data, - collaborate with colleagues, - approve decisions, - run processes, …I open an application. Each app comes with: - predefined workflows - rigid interfaces - opinionated data models - role constraints In short: software dictates how work happens. What if the application layer disappears? With AI agents, the interaction model changes: You don’t go to software. Software comes to you. Instead of navigating systems, you simply say: “Analyze last quarter’s trial performance, compare sites, draft actions, and align with the clinical ops team.” - No dashboards. - No modules. - No workflow configuration. - Just intent → execution. The emerging architecture may look very different: - Trusted, high-quality enterprise data layer - Strong governance and permissioning fabric - AI agent layer interpreting intent - Dynamic workflows generated in real time In this world: - UI becomes conversational - Workflows become adaptive - Collaboration becomes ambient - Applications become invisible The AI layer orchestrates everything. SaaS solved distribution. AI may eliminate the need for applications altogether. Traditional applications optimized for standardization. AI optimizes for personalization at scale. Instead of one workflow for everyone: Every employee gets their own adaptive operating environment. The uncomfortable implication: Many enterprise applications today may become: - Structured databases with legacy interfaces attached. - The long-term value may shift away from: screens, menus, and workflow engines …toward data quality, trust, and governance. The real future question isn’t: “Which software will win?”It may be: Who owns the trusted data layer and the governing AI layer? Because whoever controls that… controls the enterprise operating system. We may be witnessing the transition from: Application-centric enterprises → Intelligence-centric enterprises. Software may not disappear tomorrow. But the idea of applications as the primary way humans work? That assumption now looks increasingly fragile. Curious how others see this: Are SaaS applications evolving — or slowly becoming abstraction layers waiting to be replaced by AI agents? #AI #EnterpriseAI #GenAI #FutureOfWork #DigitalTransformation #Software #AgenticAI