Why AI Is Not the End of SaaS, but Its Renaissance

The CEO of Microsoft, Satya Nadella, recently made a bold statement: that AI could signal the end of SaaS as we know it. This was clearly designed to provoke a conversation—not deliver a final verdict. He wasn’t declaring SaaS dead. He was pointing to a fundamental shift in how we interact with software. 

Throughout history, every major technological shift—whether it was the move to the cloud, the rise of mobile, or the explosion of APIs—hasn’t erased what came before. It transformed it.

If you zoom in on what’s happening across the B2B SaaS ecosystem, it’s not immediately obvious whether we’re seeing the end—or just the beginning of something new. Will AI push SaaS to the sidelines, or will it elevate it to a smarter, more capable version of itself?

The argument for the “end of SaaS”

A shift from SaaS to AI-native interfaces

The core of the argument Nadella made on the BG2 Podcast: AI is becoming the primary interface between users and data. Instead of clicking through apps or navigating dashboards, users are increasingly asking a chatbot to do it for them. And in many cases, it works.

People are skipping the traditional GUI and asking tools like Copilot or ChatGPT questions in plain English. Want to know the latest sales numbers? Just type it in. Need to pull a report from your CRM? Ask the assistant. 

What makes this possible is the combination of large language models and API access. According to Nadella, SaaS platforms are all essentially just CRUD databases (Create, Read, Update, and Delete operations), layered with business rules and visual workflows. But AI can now dynamically generate queries, interact with APIs, and present results in real time. 

Much of what users would use traditional SaaS platforms for—reporting, filtering, summarizing, even automating tasks—can now be replicated without needing the app itself. This shift in user behavior raises a big question—if AI is now the front door to data and functionality, is there a point in using all the individual SaaS tools that used to live there? 

SaaS becoming interchangeable

That brings us to the common critique that SaaS tools risk becoming AI APIs packaged in slightly different ways. If every platform is just a different skin on the same language model, is there really any differentiation?

But there’s a catch. As Shaun Clowes has pointed out on the Lenny Podcast:

Surface-level interfaces are only one piece of the value puzzle of SaaS. Real enterprise software goes far deeper. 

The real value is in how tools structure workflows, manage permissions, enforce domain-specific logic, and handle complex integrations. You might be able to replace the front end with AI, sure—but you still need a robust engine running underneath.

General-purpose agents reduce the need for multiple tools

One of the more compelling arguments for why AI could make SaaS obsolete lies in the emergence of general-purpose AI agents. These agents are being trained to perform cross-functional workflows—like summarizing documents, planning and executing open-ended research projects, generating performance reports, triggering follow-up actions across tools, or even writing internal documentation. 

Instead of toggling between five different SaaS platforms to complete a task, a user can rely on a single AI interface to do it all. These agents can pull data from one tool, analyze it, format it, and take the necessary next step—all in one continuous flow. 

As AI becomes more capable of orchestrating these kinds of workflows independently, the perceived value of individual SaaS solutions starts to erode. The AI layer becomes the operating system of work, coordinating tasks across the stack, reducing friction, and boosting productivity. 

Why this narrative is incomplete

It’s true that AI is changing how we use software. People are starting to talk to tools instead of clicking through menus and dashboards. But that doesn’t mean we’re getting rid of SaaS entirely.

What’s really happening is this: the way we interact with software is becoming simpler. But behind the scenes, the software still needs to handle complex tasks—like billing rules, user onboarding workflows, security permissions, and integrations with other tools. 

These backend systems are actually becoming even more important because they carry the logic and structure that make the business work.

In other words, the narrative around the “end of SaaS” misses the bigger picture. AI isn’t killing SaaS. It’s reshaping what SaaS can—and should—do.

The real shift: SaaS, rewired by AI

SaaS is evolving, not disappearing

Let’s reframe the question: What if AI isn’t the replacement for SaaS, but the next step in its evolution? Instead of becoming obsolete, SaaS platforms are absorbing AI and using it to become more intelligent, more predictive, and more responsive. 

The numbers back this up. The AI SaaS market was valued at $71.54 billion in 2024 and is projected to reach a staggering $775.44 billion by 2031. That might just be a projection, and even Satya Nadella has pointed out that for all the hype, we haven’t yet seen a measurable impact from AI on global GDP. The value is still largely potential, not proof.

But what that tells us is important: we’re still early. The shift is underway, and forward-thinking SaaS platforms are already laying the groundwork by embedding AI as a core part of how they deliver value. 

Think of SaaS as the foundation. AI sits on top of it, turning it from a static system of record into a living, breathing engine of insight. AI doesn’t need to replace SaaS—it enhances it.

AI as the operating system, not the product

What’s emerging now is a new model where AI functions like an operating system. It runs quietly in the background, embedded into each workflow and data stream. Users might interact with a chatbot, but the logic that determines what happens next still comes from the SaaS platform.

This model moves away from flashy front-end tools and toward deeply integrated intelligence. And it also shifts focus away from general-purpose AI to domain-specific, contextual AI. It’s not about having a chatbot. It’s about having a chatbot that knows your product, your customers, your rules. 

SaaS UI still matters—and AI makes it better

While natural language interfaces are impressive, they’re not always the best or fastest way to interact with software. A well-designed UI still has real advantages, especially for repeatable tasks and quick data access. 

For many users, clicking through filters, dropdowns, or dashboards is more efficient than prompting an AI assistant with something like, “Show me all the calls I did today,” then following up with, “Include the time per call and the expected deal value.” 

Chat interfaces—especially in agent-based AI tools—can also become frustrating in more complex scenarios. They lack visibility into workflows, make editing or reversing decisions harder, and it can be difficult to coordinate between multiple AI agents and human users in a linear chat format. 

It’s not just about speed—it’s about preference. Some users feel more in control with visual tools, while others appreciate the flexibility of a conversational interface. 

That’s exactly why SaaS doesn’t need to be replaced by AI—it needs to be enhanced by it. The future lies in choice: platforms that offer both intuitive UIs and powerful AI layers will serve a wider range of user styles and workflows. 

As Rob Walling puts it,

AI can surface recommendations, automate repetitive steps, and answer ambiguous questions, while the UI continues to offer clarity, structure, and speed for those who prefer a more hands-on approach. Together, they make SaaS more adaptable—and more intelligent—than it’s ever been.

The SaaS solution: Unification, not fragmentation

The argument that AI agents reduce the need for multiple SaaS tools is valid—because the current SaaS stack in many businesses is fragmented. One tool for CRM, another for support, another for product analytics, and yet another for customer onboarding. It's no wonder an AI agent that works across all of them sounds more efficient.

But the solution isn’t to get rid of SaaS—it’s to rethink how SaaS is delivered.

Modern SaaS platforms are evolving from single-purpose tools into integrated ecosystems. The most forward-thinking platforms are unifying use cases that previously lived in separate apps. 

Even better, many SaaS platforms are embedding AI agent functionality within the product itself. So instead of needing an external AI tool to interact across systems, users can ask questions, trigger workflows, or get recommendations directly inside the platform. It’s a shift from “best-of-breed” collections of tools to purpose-built, intelligent operating environments.

By consolidating key workflows and layering AI inside the product, SaaS tools solve a problem agents were supposed to address: too many tools, not enough context. They offer a controlled, secure, and deeply integrated experience that feels like an agent—but with the governance and reliability enterprises require.

Workflow enablers become decision engines

We’re also seeing a functional evolution. SaaS platforms used to be mostly about facilitating tasks—logging tickets, storing customer info, pushing data between systems. Now they’re becoming active participants in decision-making.

Here’s what that looks like in practice:

• Anomaly detection in Customer Success: Spotting unusual activity across the customer lifecycle without waiting for someone to notice manually.
• Dynamic pricing in eCommerce platforms: Adjusting product prices based on competitor activity, demand forecasts, and inventory levels using machine learning.
• Intelligent document processing in HR tools: Scanning resumes, extracting key qualifications, and shortlisting candidates who best match the job description.
• Proactive maintenance alerts in SaaS for hardware/IoT management: Notifying users of potential equipment failure based on usage data and sensor readings.

These aren’t bolt-ons. They’re native capabilities that change how SaaS tools create value.

SaaS platforms are evolving from passive systems into active decision engines. It’s no longer enough to just collect data—teams need to know what to do next. At Velaris, we’ve built our platform around this exact idea. We bring together signals from product usage, support interactions, CRM notes, emails, and call transcripts—not just to store them, but to interpret and act on them.

Instead of waiting for a Customer Success Manager to dig through dashboards, Velaris surfaces insights automatically. Think nuanced churn prediction that analyzes ticket trends and email tone, or smart stakeholder alerts that highlight potential upsell opportunities based on both structured data and conversational signals. 

This shift reflects what many leading SaaS companies are doing today: embedding intelligence directly into the product experience—not as a layer on top, but as a core function.

Why orchestration is still the moat

The reason SaaS continues to matter is because it provides orchestration: the logic, workflows, triggers, and integrations that make things move from insight to action.

So even as AI adds new capabilities, it’s the orchestration layer that determines how those capabilities are used—and how effective they are.

And that’s exactly where we’re focused at Velaris. We don’t see AI as something that replaces the entire platform—we see it as an opportunity to rethink how a Customer Success platform should function. Our goal isn’t to automate everything. It’s to deliver the kind of intelligence that makes teams more responsive, more proactive, and more effective.

The broader B2B SaaS implication

Enterprise buyers want outcomes, not models

The flood of new AI tools has made one thing very clear: AI itself isn’t the product. It’s the plumbing. What businesses want are outcomes—faster onboarding, reduced churn, higher engagement. They don’t want to configure another AI dashboard. They want to see a workflow that works.

This is why SaaS platforms still matter. They turn raw AI infrastructure into usable, repeatable, scalable workflows. The underlying model might be shared across platforms—but the implementation, the context, and the results are what set products apart.

The future stack: Model + workflow + context

Going forward, the winning formula in B2B SaaS won’t just be about model access. It’ll be about combining three things:

1. Model: An AI engine capable of understanding and processing complex data.

2. Workflow: A structured set of actions and automations tied to real business processes.

3. Context: Deep, domain-specific data and user signals that guide the model.

Generic chatbots won’t cut it. SaaS platforms that win will be those that can marry proprietary data with embedded workflows. They’ll serve users not just by answering questions, but by knowing which ones to ask and when to act.

What this means for SaaS companies

If you’re building SaaS today, it’s no longer enough to just “add AI.” The question is: How deeply is AI embedded into your workflows? How context-aware is it? Can it adapt based on behavior, segment, lifecycle stage?

At Velaris, we’ve found that the real opportunity isn’t in replacing SaaS with AI—it’s in reimagining what a SaaS platform can do when AI is embedded at its core. We’re not just adding AI features—we’re rebuilding the foundation to make intelligence a native part of every workflow.

B2B organizations today are drowning in data, yet struggling to act on it. That’s because the most important insights are scattered across disconnected systems—CRM entries, product analytics dashboards, support tickets, customer calls, and message threads. 

We designed Velaris to change that. By integrating deeply with the tools teams already use—Salesforce, Stripe, Slack, Zendesk, and more—we centralize all customer data in one place. Structured data like usage frequency and license consumption sits alongside unstructured inputs like email tone, meeting transcripts, and call notes, with both being easily visible.

This is one of the biggest advantages we’ve built into Velaris—it collects and organizes data in the right place. While AI is great for retrieving specific pieces of information, they often fall short when it comes to answering more complex, cross-system questions. That’s because some insights require data from multiple tools to be stored and interpreted together, not just pulled separately in real time.

For example, you might want to know: “Which high-value customers submitted tickets about the new feature, had a drop in usage, and haven't been contacted by their CSM in the last 10 days?”

That’s where Velaris has the edge. We centralize structured and unstructured data—so AI can actually ask meaningful questions across those sources. 

Want to generate a stakeholder map for an enterprise account based on email activity and meeting transcripts? Done. Want to surface signals of frustration across your highest-value accounts and route them to the right success manager? You don’t need to go digging—Velaris AI flags it. 

This is how we see the future of SaaS—not just tools for managing workflows, but platforms that think, learn, and help you act faster and smarter. We’re building Velaris to be a one of a kind platform: one that uses AI not to replace people in SaaS, but to amplify them. 

Conclusion

So no, SaaS isn’t over. It’s entering a new chapter. One where the interface may look different—fewer dashboards, more conversations—but the core value of SaaS platforms remains the same: turning complexity into clarity, and data into direction.

In a world where businesses are swimming in raw data and generative content, the tools that help people make sense of it all will only become more important. 

Not all SaaS tools will adapt equally. Narrow, single-purpose products that don’t offer integrated workflows or deep context may be more vulnerable to being replaced by flexible, AI-native solutions. But platforms like Velaris, which understand not just what’s happening but what needs to happen next, are well-positioned to lead.

This isn’t the end of SaaS. It’s a renaissance.