Why the pivot away from SaaS pricing is a governance question, not just a product one
The shift under way
For two decades, software valuations rested on a simple formula: recurring subscription revenue, high gross margins, and predictable renewal rates. That formula is now under genuine pressure. As AI agents increasingly perform the work that human users used to do inside software products, the per-seat licence that funded the SaaS era is losing its logic. If the software is doing the job rather than assisting a person to do it, charging by the seat no longer reflects where the value sits.
The industry has coined a distinction worth boards understanding clearly. AI-enabled software bolts a generative layer onto an existing product, a chatbot here, a summarisation feature there. AI-native software is built so that the AI is the product itself; remove the model and there is nothing left to sell. This is not a cosmetic difference. It changes how a company should be priced, resourced, and governed.
The models emerging to replace SaaS
Boards evaluating a pivot, or assessing a portfolio company that claims to have made one, should expect to see revenue built on some combination of the following:
– Usage-based pricing charges by query, task, or compute consumed, rather than by user licence. This tracks value delivered more accurately, but it also transfers cost volatility onto the vendor. Margins move with model inference costs, not just with customer count.
– Outcome-based pricing goes a step further, charging for a resolved ticket, a completed reconciliation, or a booked appointment. This aligns price directly with value, which investors like. It also requires the vendor to measure and stand behind outcomes reliably, which is a harder operational and legal commitment than shipping features.
– Labour substitution pricing frames the AI product against the cost of the role it replaces, rather than against a software budget line. This is a materially different sales conversation, often moving the buying decision from an IT budget to an operations or workforce budget, with different approval chains and different scrutiny.
– Infrastructure and platform positioning sees some incumbents reposition themselves as the substrate other AI agents and applications run on top of, monetising through API and platform fees rather than end-user subscriptions.
Why this is harder than it looks
The economics of AI-native businesses are not simply a more efficient version of SaaS economics. Early-stage AI-native companies are running gross margins closer to 25 percent, against 75 percent or higher for mature SaaS businesses, because inference costs scale directly with usage in a way seat-based software never did. Retention patterns look different too, with gross retention in AI-native cohorts trending well below the 80 percent-plus boards are used to seeing from enterprise SaaS. Neither of these facts is disqualifying, but both should be sitting on a board’s risk register before the growth story is accepted at face value.
There is also a sequencing problem specific to incumbents attempting the pivot. A legacy SaaS company moving to AI-native must rebuild its product architecture, replace the pricing model that generated its existing revenue, and do both while the public and private markets are already pricing in the assumption that the transition is complete. That is three transitions running concurrently, and most organisations are not structured to manage more than one at a time. Directors should treat management confidence about timeline with the same scepticism they would apply to any major transformation program, because the pattern of underestimation here is well established.
What buyers are reporting
The market signal is more contested than the more breathless commentary suggests. Recent CIO surveys show a substantial share of enterprise buyers, in the order of half, saying they are ready to replace incumbent software vendors with AI-native alternatives. At the same time, enterprise buyer surveys from early 2026 found the clear majority expect their existing vendors to benefit from generative AI rather than be displaced by it, with only a small minority expecting incumbents to lose out altogether. Buyers, in other words, are showing a preference for evolution over replacement, provided the incumbent executes. That preference is not a guarantee, and it is contingent on execution quality, but it does temper the more absolute “SaaS is dead” narrative circulating in parts of the investment community.
Questions for the boardroom
Directors overseeing a proposed pivot, or assessing an AI-native target for investment, should be pressing management on a small number of concrete points: how outcomes are defined and measured for pricing purposes; what happens to gross margin as usage scales rather than seats; who owns accountability when an autonomous agent produces a poor or harmful outcome; and whether the transition plan sequences architecture, pricing, and market communication realistically, rather than assuming all three can move together without strain.
The business model question is ultimately a governance question. Getting the pricing architecture wrong is not just a commercial misstep, it is a failure of the strategic oversight boards exist to provide.
About: Gary Morgan is a director, board advisor and principal consultant at MPT Innovation Group, specialising in governance, technology strategy, and organisational transformation for private and not-for-profit organisations. He is a Fellow and Member of the Queensland State Council of the Governance Institute of Australia, and an Adjunct Fellow and Member of the Griffith University Industry Advisory Board for the ICT School. Gary publishes regularly on board governance, AI, technology, and cybersecurity.
Acknowledgment: This article represents the author’s independent views and incorporates AI-assisted research and drafting.
Sources:
Appscrip. (2026). The Future Of SaaS 2026: What’s Actually Changing
Vendasta. (2026). AI-Native SaaS in 2026: How to Catch Up Without Rebuilding From Scratch
BetterCloud. (2026). AI and the SaaS industry in 2026
SaaSmag. (2026). The Rise of AI-Native SaaS: Born-AI Companies Scale Faster
Chargebee. (2026). 2026’s Real SaaS Threat Isn’t AI. It’s Business Model Debt State of Brand. (2026). Can a SaaS Company That Retrofits AI Ever Be Seen as AI-Native?
English