It is January 2026, and the software sector is facing its most existential test in over a decade.
For years, the “Rule of 40,” recurring revenue, and high switching costs made SaaS the darling of the venture capital asset class. But the start of 2026 has brought a brutal reality check. A basket of SaaS stocks tracked by Morgan Stanley is off to its worst start since 2022, with valuation multiples crashing to record lows of 18 times projected earnings, a far cry from the 55x average of the past decade. The trigger? Not interest rates, but relevance.
The recent release of Anthropic’s “Claude Cowork” has rekindled deep-seated fears that AI agents are not just features… they are replacements for the software itself. As we look at the wreckage in the public markets, where stalwarts like Intuit and Salesforce have tumbled double digits, the message for founders is clear: the playbook that built the last decade’s unicorns is being rewritten in real-time.
Here is how AI is disrupting the SaaS model and how founders must adapt to survive.
The “Per-Seat” Model is Under Siege
The traditional investment thesis for SaaS relied on the predictability of subscription revenue based on human headcounts. That narrative is collapsing now that investors fear AI agents running 24/7 will replace the human workers who used to occupy those paid seats.
For founders, this signals an urgent need to rethink business models. As portfolio manager Bryan Wong notes, it is difficult to assign a valuation multiple to a company “if they’re going up against AI agents that… have the ability to complete tasks, with big projects getting done in a day”. If your revenue scales linearly with human users, you are vulnerable. The next generation of successful startups will likely shift toward outcome-based pricing or consumption models, capturing value from the work generated by the AI agent rather than the human staring at the dashboard.
Code is a Commodity; Domain Knowledge is the Moat
With AI writing code and designing interfaces instantly, technical barriers to entry are evaporating. However, the market offers a clue on where safety lies.
Orlando Bravo, of private equity giant Thoma Bravo, argues that the recent sell-off is a buying opportunity for specific types of companies: those with deep domain knowledge. “Software is not at all about the code… Software is about your domain knowledge,” Bravo notes, suggesting that specialized verticals like payroll or cybersecurity remain insulated because accuracy is non-negotiable and processes are complex.
Founders should stop selling “software” and start selling deep vertical expertise. A generic CRM is vulnerable; a highly specialised compliance platform for a regulated industry where the “franchise value” lies in understanding a specific process better than anyone else is defensible. I wrote a bit more about this in my recent post on vertical AI.
The Great Divergence: Infrastructure vs. Application
The “Magnificent Seven” trade has fractured. While software stocks sink, chipmakers like Nvidia and hyperscalers (Amazon, Microsoft, Alphabet) continue to outperform because they provide the infrastructure for the AI build-out. The market currently has greater certainty about the companies powering AI than the software companies integrating it.
Incumbents like Salesforce and Adobe have touted their AI features (like Agentforce), but investors remain skeptical as these features haven’t significantly moved the revenue needle yet. Mere integration of generative AI features is no longer a differentiator; it is table stakes. To command a premium, founders must demonstrate that their AI implementation drives clear, revenue-generating traction, distinct from legacy efficiency gains.
Growth is the Only Metric That Matters
In this new environment, the market is punishing low growth ruthlessly. Data shows that revenue growth is currently a better predictor of returns than margins or profitability. Companies growing faster than 20% (like Palantir) are seeing stock appreciation, while those growing below that mark are being sold off.
As Tomasz Tunguz points out, “In 2016, investors questioned valuations. In 2026, they question relevance”. Founders cannot cut their way to a high valuation in 2026. Efficiency is important, but if customers aren’t expanding their usage, proving your product remains relevant in an AI-first world, no amount of profitability will save you from a valuation crush.
Pivot to Autonomous Outcomes
The SaaS playbook of the last decade is being rewritten. The fear gripping the market is that AI doesn’t just lower costs, but that it “exposes products that were convenience, not necessity”.
The era of selling seats is ending; it’s not going to change right away, but the era of selling autonomous outcomes has begun. Clients will start paying for results rather than tools. Founders who can transition from building tools for humans to building platforms that deliver finished work will define this new cycle.
At Remagine Ventures, we’re looking at Israeli pre-seed opportunities that are embracing this shift. If you’re building something relevant, lets chat!
I'm a former general partner at google ventures, head of Google for Entrepreneurs in Europe and founding head of Campus London, Google's first physical hub for startups.
I'm also the founder of Techbikers, a non-profit bringing together the startup ecosystem on cycling challenges in support of Room to Read. Since inception in 2012 we've built 11 schools and 50 libraries in the developing world.
- Weekly Firgun Newsletter – February 27, 2026 - February 27, 2026
- The Rise of the Personal AI Agent - February 26, 2026
- Pre-Seed Is Stronger than ever, but less startups are getting funded - February 25, 2026
