The State of SaaS Disruption 2025 | Part 4

The Investors Playbook: How to Bet on the New SaaS Landscape

In the first three parts of this series, we’ve established a new reality: the SaaS industry is being fundamentally rewired by AI. The old moats are eroding, Agentic AI is threatening the user interface, and a great bifurcation is splitting the market between the vulnerable and the resilient.

For investors, this means the playbook that generated historic returns over the last decade is now obsolete. Traditional SaaS metrics—while still useful—are no longer sufficient to gauge long-term viability. ARR growth is vanity if your product can be replicated by a lean, AI-native competitor in six months. A healthy LTV/CAC is a lagging indicator if your customers can soon achieve the same outcomes without your software.

As an investor and advisor, my focus has shifted from evaluating a company’s current trajectory to assessing its structural resilience in the face of this AI tsunami. Investing in SaaS in 2025 requires a new thesis, a new set of questions, and a disciplined ability to distinguish durable value from sophisticated hype.

This is the playbook for placing bets in the new SaaS landscape.

Key Takeaways for Investors

Your Old SaaS Scorecard is Obsolete. Stop leading with seat-based growth and ARR multiples. Start with data strategy, workflow entrenchment, and business model adaptability. The game has changed from capturing users to capturing unique value.
Scrutinize the Data Moat Above All Else. In an era of commoditized code, proprietary data is one of the few truly defensible assets. Does the company have a “data flywheel” that creates a compounding advantage? Or is its data strategy an afterthought?
Distinguish “AI-Native” from “AI-Washing.” Many incumbents are “bolt-on” AI features as a marketing tactic. You must learn to spot the difference. True AI-native solutions have AI embedded in their core architecture and value proposition, not just layered on top.
Bet on Teams with AI Literacy, Not Just AI Experts. A siloed “AI team” is a red flag. The winning companies will be those that foster AI literacy and an experimental mindset across the entire organization, from product to sales.

The New Thesis: From Growth Metrics to Moat Durability

For years, the venture capital community has evaluated SaaS companies on a relatively standard set of metrics. We looked for predictable, recurring revenue, high gross margins, and efficient customer acquisition. These metrics measured the health of the business model. They did not, however, always measure the durability of the underlying value proposition.

The AI revolution has exposed this gap. The central question for an investor is no longer just “Is this a good business?” but “Does this business have a right to exist in a world where AI can do its job?”

This forces a shift in due diligence, prioritizing qualitative, strategic factors over purely quantitative, backward-looking metrics. As leading VCs like Andreessen Horowitz and Sequoia have noted, the biggest opportunities will be in companies that are not just using AI, but are fundamentally re-imagined by it. This means looking for businesses that fit a new profile of success.

The AI Due Diligence Checklist: A Framework for Evaluation

To put this new thesis into practice, I use a rigorous checklist designed to probe for AI-era defensibility. These are the questions that should be front and center in every pitch meeting and due diligence process.

1. AI Strategy & Vision

Is AI a core, integrated part of the product vision, or is it a “bolt-on” feature set designed to signal innovation?
How, specifically, does the company’s use of AI solve a customer’s pain point in a way that was previously impossible? Can they quantify the value?
What is the long-term vision for autonomy? Are they building towards an Agent-as-a-Service (AaaS) model, or are they content with being a human-operated tool?

2. Data Moat & Strategy

Does the company have access to proprietary, high-quality datasets that are difficult or impossible for competitors to acquire?
Is there a “data flywheel”? Does more usage of the product generate more unique data, which in turn improves the AI, creating a compounding advantage?
How is data governed and secured? Is the company prepared for the ethical and regulatory complexities of training models on customer data?

3. Technology & Architecture

Is the tech stack “AI-ready”? Can it support demanding AI workloads, or is it a legacy monolith that will require a painful and expensive refactor?
Do they have robust, well-documented APIs? How are they preparing for a “headless” future where agents are their primary users?
What is their model strategy? Are they building proprietary models, fine-tuning open-source models, or simply wrapping a third-party API? What are the associated costs, dependencies, and risks?

4. Team, Talent & Culture

Does the leadership and technical team possess deep AI/ML expertise, or are they learning on the fly?
Is AI expertise embedded within the product teams, or is it siloed in a separate “AI lab”? The former is a sign of true integration; the latter is often a sign of “innovation theater.”
Is there a culture of rapid experimentation and data-driven decision-making across the entire organization?

5. Business Model Adaptability

How does the company plan to price and monetize the value created by AI? Are they exploring outcome-based or consumption-based models?
How will a shift away from per-seat pricing affect their financial model and sales strategy? Are they prepared for potential revenue contraction in some segments due to AI-driven productivity?

Signal vs. Noise: How to Spot Superficial AI-Washing

One of the biggest challenges for investors today is cutting through the noise. Every SaaS company now claims to be an “AI company.” Distinguishing genuine innovation from marketing buzz is critical.

Red Flags (AI-Washing):

AI is primarily discussed in marketing materials, not in the product architecture.
The use case is a generic “AI assistant” or chatbot with no deep workflow integration.
The team cannot articulate a clear data strategy or explain how their AI creates a unique competitive advantage.
The solution is a “thin wrapper” around a public AI model with no proprietary technology or data.

Green Flags (AI-Native Strength):

AI is core to the founding vision and solves a problem in a fundamentally new way.
The company has a clear strategy for acquiring a unique, defensible data asset.
The business model is being rethought around AI, with experiments in value-based pricing.
The team demonstrates deep AI literacy at all levels of the organization.

The massive capital flowing into AI infrastructure—with hyperscalers prioritizing GPUs for AI over CPUs for traditional cloud computing—is the ultimate tell. The platform shift is real. As investors, our job is to find the companies that are building on the new platform, not those left defending the old one. This requires discipline, a new framework for evaluation, and the courage to bet on the seismic shifts that are redefining an entire industry.

Coming up in Part 5: We’ll turn our focus to the builders. In The Challenger’s Advantage: A Blueprint for Startups and Cloud-Native Scale-Ups, I will outline the playbook for those best positioned to win in this new era: the agile, unburdened disruptors.