We are living through simultaneous SaaS repricing and re-platforming trends. This has significant consequences for public valuations and private portfolios.
In the post ZIRP-era, public markets continue to reward growth but also started paying much more attention to cash generation. Software companies are no longer rewarded for growth alone. Today, they need to show both growth and efficiency. That said, markets still pay double-digit multiples of top line for high growth. As our friends at Meritech Capital highlight, markets still reward growth more than cash flow. The median EV/Implied ARR multiple for a 20-30% grower with 10-20% cashflow margin is significantly higher than that for a sub 10% grower with over 30% free cash flow margin.
Source:Meritech Software Pulse, 3/20/2026
In our view, the main issue is that the majority of public SaaS companies are simply not growing that fast anymore, mostly up to 20%, and, more importantly, the indicators of durable, future growth are also not as encouraging as they used to be. For example, currently the median net dollar retention (NDR) is sub 110%, which is well below the ~118% range at peak ZIRP. Notably, this is well below the 120%+ NDRs, which the fastest growing companies (defined as annual growth of 25%+) are maintaining in the current market.
IT budgets’ growth also remains unexciting. Per the 4Q 2025 Morgan Stanley CIO survey, spending is projected to grow +3.4% in 2026 vs. +3.5% in 2025; remaining below the 10-year survey average of +4.1%. The CIOs focus for 2026 is on AI/ML, security software, digital transformation and ERP & CRM applications, which are all expected to see much higher spending growth with the top category (AI/ML) at low double digits. Therefore, we are observing a relatively rational repricing of businesses which were still priced, to some extent, on the hope of growth reacceleration post a cycle reset and with a potential AI catalyst; so far, this reacceleration has not materialized.
While AI has driven efficiencies (revenue across software companies is now higher per FTE), so far there is little evidence it has driven meaningful top line growth acceleration for the majority of public software companies. Even the software companies deemed beneficiaries of AI tailwinds, e.g. Datadog, Snowflake, etc., which saw their stock prices decouple vs. the rest of software in the last 12 months, have seen some of that gain erased in early 2026.
Public software stocks are now also suffering from investors’ increasing fears that instead of growth acceleration AI might prove to be destructive to the terminal value of most of the software application layer companies as they get disrupted by innovators.
From what we have come across, AI is breaking the traditional SaaS moats in at least three major ways:
- Software is cheaper to produce and platforms are being built faster: When the marginal cost of shipping “one more feature” collapses, feature-based moats collapse with it. Your competitor doesn’t need as many engineers to catch up; they need taste, speed, and a strong loop. AI also makes it easy to slice a workflow into micro-products driving fast adoption and easy to re-bundle them later into a platform driving fast commoditization.
- The traditional SaaS UI and monetization model is no longer relevant: AI is changing what software is, shifting it from tools used by humans, to integrated systems of action. We are entering an agentic era in which both system architecture, as well as the monetization model, are rapidly evolving. If an agent can complete a task end-to-end, a traditional interface becomes optional. SaaS that survives will evolve its UI and UX and will price on impact/ usage and not on a per seat (perhaps with few exceptions).
- Private, AI-first companies can run faster with less attractive unit economics: Public software companies face a difficult balancing act. They have the hard task of simultaneously re-engineering their products and vision, demonstrating meaningful growth acceleration from AI, and defending their security & compliance promise and margins. By contrast, AI-native, private companies are able to grab market share fast and post exceptional top line growth without the challenge of doing so at SaaS-like margins. Many operate at significantly lower margins, often in the 20-40% range, and it remains uncertain where these will settle in the mid-term.; They also do not face the self-disruption dilemma.
The best pre-ChatGPT software companies, public or private, are reinventing themselves fast. Those that succeed will likely end up being bigger, more durable, and more valuable. But beyond the hyperscalers, public markets suggest they still do not know which ones will be successful and are pricing the risk indiscriminately.
We believe that the companies that don’t evolve through this re-platforming period, will die or end up as sub-scale / non-VC-like outcomes (as it relates to private, legacy portfolios). This does not mean all public/private SaaS will be commoditized and perish, nor does it mean all current AI-native “high flyers” will grow into winners once they mature. As in prior cycles, private market returns will be driven by vintage diversification, prudent portfolio construction, and access to the best managers and n-of-1 founders that can adapt fast to change, execute relentlessly, and recognize trends early.
Past experience has shown us that dislocated markets offer the best opportunity to generate alpha, so happy hunting!
