The State of SaaS Disruption 2025 | Part 2

The Agentic Shift: Why the Future of Software Has No UI

In the first part of this series, I described the AI Tsunami as two distinct waves. The first, Generative AI, is an evolutionary force that accelerates the existing SaaS paradigm. The second wave is the one we must now turn our full attention to. It is a revolutionary force with the power to fundamentally re-architect the entire software industry.

This is the Agentic Shift.

If GenAI changed how we create and interact with content and code, Agentic AI changes what software is and how it delivers value. It represents a move away from software as a passive tool that humans operate, towards software as an autonomous colleague that executes complex tasks on our behalf. As an industry, we have spent decades perfecting the Graphical User Interface (GUI). The agentic shift threatens to make it irrelevant.

Understanding this paradigm is not optional. It is the single most important strategic consideration for any software leader or investor for the remainder of this decade.

Key Takeaways for the C-Suite & Investors

• AI Agents Are Not Chatbots. An AI Agent is an autonomous system that can understand a goal, create a multi-step plan, and execute it across different applications. It moves from answering queries to achieving outcomes.
• The “No UI” Future is Metaphorical, But the Threat is Real. The prediction is not that all GUIs will literally vanish. It’s that the primary point of interaction will shift from clicking through an application to issuing a natural language command to an agent. This makes the UI a secondary, and less valuable, asset.
• Your SaaS Product Risks Becoming a “Headless” Commodity. In an agent-driven world, your application’s value may be reduced to its underlying logic and data, accessed via an API. Without a direct user relationship, brand loyalty and pricing power erode dramatically.
• A New Business Model is Emerging: Agent-as-a-Service (AaaS). The market is shifting from selling access to a tool (SaaS) to selling a guaranteed outcome delivered by an agent (AaaS). This requires a complete rethink of pricing, value propositions, and go-to-market strategies.

From Assistant to Agent: The Leap in Capability

For years, we’ve had AI “assistants” like Siri and Alexa. They are powerful but fundamentally reactive. They respond to direct, simple commands within a limited context. An AI Agent is a different class of technology entirely.

As I outlined in my research, an agent possesses three core capabilities that set it apart:

1. Memory: It maintains both short-term context for the task at hand and long-term knowledge to learn and improve over time.
2. Planning: It can decompose a high-level goal (e.g., “increase sales in the EMEA region”) into a sequence of logical sub-tasks.
3. Action: It can interact with the digital world to execute those tasks—calling APIs, accessing databases, and even using other software tools.

Consider the practical difference. You tell a SaaS application: “Show me my sales dashboard.” You then manually analyze the data, formulate a plan, and execute it in other applications.

You tell an AI Agent: “Find my top three underperforming sales reps in Europe, identify the root cause by analyzing their call logs and pipeline data, and draft a performance improvement plan for each.”

The agent does the work. This is not science fiction; it is what top technology firms are building towards right now. Gartner has named Agentic AI the number one strategic tech trend for 2025, predicting that agents will autonomously handle 80% of standard customer service requests by 2029. This is the new benchmark for software intelligence.

The No UI Paradigm: What Happens When Your Interface Isnt Yours?

For decades, the UI has been the heart of a SaaS product. It’s where we invested millions in R&D and design, and it’s how we built user habits and brand loyalty. The agentic shift threatens to completely unbundle the user experience from the application itself.

Microsoft CEO Satya Nadella has described this as the emergence of a new “AI tier” in the software stack. This layer, powered by agents, sits between the user and the application layer. Instead of logging into your CRM, the user simply tells their agent what they need. The agent then interacts with your CRM’s backend via its API.

In this scenario, your SaaS product becomes “headless.”

Your beautifully designed dashboards, intuitive forms, and carefully crafted user flows are bypassed. The user’s experience is with the agent, not with your product. This has devastating implications for defensibility:

• Erosion of Stickiness: User habits are tied to the agent, not your app.
• Intensified Competition: If multiple SaaS products offer similar functionality via APIs, an agent could theoretically swap between them based on price or performance, turning your service into a commodity.
• Loss of Brand Identity: Your brand is relegated to a line item in an agent’s execution log.

This doesn’t mean GUIs will disappear overnight. As I’ve seen in every major technology transition, from mainframe to cloud, there will be a long period of hybrid use. Users will still need visual interfaces for complex data analysis, configuration, and oversight. However, the balance of power will shift. The primary value will migrate from the interface to the underlying data, logic, and outcomes.

The Rise of Agent-as-a-Service (AaaS)

If the primary value of software is shifting to autonomous execution, then the business model must evolve accordingly. This is where Agent-as-a-Service (AaaS) comes into play.

Instead of buying a license for a project management tool, a company might subscribe to a “Project Management Agent” tasked with ensuring projects are delivered on time and within budget. The pricing model shifts from a per-seat fee to a fee based on the outcomes achieved or the complexity of the tasks managed.

This is a far more scalable and valuable model. It aligns the software’s price directly with the economic value it creates, a goal the SaaS industry has chased for years. However, succeeding as an AaaS provider requires a fundamentally different mindset and architecture. You are no longer selling a tool; you are selling a trusted, reliable, autonomous service.

The Challenges Are Real, But the Direction is Clear

The vision of a fully agentic future faces significant hurdles. The quality and structure of enterprise data remain a mess. Building trust in autonomous systems for mission-critical operations will take time. Real-world workflows are often more nuanced and complex than current agents can handle.

Because of this, the transition will be gradual. Agents will start by augmenting human capabilities, then automate more defined tasks, and eventually orchestrate complex, cross-functional processes. But make no mistake: the direction of travel is clear.

As a leader, your strategic imperative is to prepare for this “headless” future now. You must begin architecting your products to be “agent-native,” with robust, well-documented APIs. You must ask yourself: if my UI disappeared tomorrow, what unique value would remain?

If you don’t have a good answer, you are building on borrowed time.

Coming up in Part 3: I will analyze the impact of these forces on the market. In Vulnerability & Resilience: Who Survives the Great SaaS Shake-Up?, we will identify which SaaS segments are most at risk and where the pockets of durable value lie.