Will Generative AI Bring Software to $0?
Transformation of SaaS startups via GenAI—embracing outcome-based pricing and navigating revenue recognition challenges
The Rise
The explosive rise of generative AI (GenAI) is reshaping the software-as-a-service (SaaS) landscape, delivering unprecedented efficiency by minimizing human involvement.
As SaaS platforms integrate GenAI, companies are pivoting from traditional seat-based or usage-based pricing to innovative outcome-based models—where customers pay only for tangible results (kind of a hurt-piece for current SaaS’s already printing MRR—but, doesn’t have to be…if you adapt!), like an AI chatbot resolving a query without escalation.
This evolution promises alignment between value delivered and costs incurred, but it introduces critical accounting complexities under ASC 606 (Revenue from Contracts with Customers).
In today’s letter, we'll explore these shifts, drawing on core accounting principles, while incorporating recent global developments from 2023–2025.
FORWARD
Wanted to reiterate the schedule moving forward. At least 3 letters per week. 1 paid letter exclusive to subscribers from release. 2 free to view letters that become locked—viewable only by paid subscribers after 14 days.
This way, you can subscribe with just your email and not have to pay a cent—and can still access a lot of great material as it is released (just not the full archive). If you are able, and have gained value from my work, consider becoming a paid subscriber today.
Recent events underscore the momentum: The global AI market surged to $391 billion in 2025, projected to hit $1.81 trillion by 2030, fueled by widespread adoption in SaaS.
PwC's 2025 AI predictions highlight how businesses are leveraging AI for transformation, with SaaS firms reporting up to 30% operational cost reductions and 25% revenue gains through GenAI integrations.
However, challenges like AI's escalating energy demands—raise sustainability concerns that could slow adoption amid “global climate pressures”—we know its bologna but still something that will slow progress.
Ethical issues and regulatory hurdles, such as the EU AI Act's enforcement rollout in 2025, add friction, potentially delaying outcome-based rollouts in privacy-sensitive sectors.
Why Outcome-Based Pricing Is Gaining Traction
Outcome-based pricing ties fees directly to results, such as successful AI-driven resolutions in customer support. This contrasts with legacy models like per-user licenses, offering flexibility in an AI-driven world where value stems from automation rather than access alone.
Benefits include stronger customer trust, as payments reflect real impact, and competitive differentiation in a market where AI-native SaaS is booming—evidenced by trends like hyper-personalization and autonomous workflows projected for 2025.
Yet, this model isn't without detractors. Recent analyses from 2023–2025 point to pitfalls: Revenue instability from attribution disputes (e.g., was the outcome purely AI-driven?), measurement complexities, and extended sales cycles that hinder predictability (How can people accurately keep their books?).
Investors often favor stable recurring revenue, and GenAI's high computational costs exacerbate these issues, as seen in critiques of traditional SaaS models struggling to adapt.
The 2024–2025 tech sector slowdown, marked by layoffs and cautious spending amid economic uncertainty, has amplified these concerns, making some firms hesitant to abandon proven pricing structures. (I tend to agree with this skepticism—building with it in mind, but still using a monthly paywall behind value-adding software.)
Determining the Nature of the Promise: Access vs. Delivery
At the heart of revenue recognition lies a pivotal question: What exactly is the SaaS provider promising? Under ASC 606, this distinction drives how revenue is measured and timed.
Promise as a Stand-Ready Obligation (Access-Focused): Here, the core commitment is continuous platform availability over a period, akin to a subscription. Customers access GenAI tools as needed, but revenue isn't tied to specific usages. Variable fees from outcomes (e.g., per successful resolution) are treated as additional consideration, often estimated upfront if predictable. This model supports ratable recognition, providing stability amid AI's variability.
Promise as Delivery of Goods or Services (Outcome-Focused): The emphasis shifts to providing a defined quantity of results, like ‘x’ number of AI-resolved queries. Each outcome counts as a distinct performance obligation, potentially triggering recognition upon delivery. If customers can opt for more, evaluate for material rights (e.g., discounts on extras) as separate obligations.
Recent examples illustrate this: In 2025, firms like those in L.E.K. Consulting's analysis are adopting value-driven models, but attribution challenges—where outcomes blend AI and human inputs—complicate classification, leading to disputes and delayed revenue.
Supporting this shift, McKinsey's 2025 workplace report notes AI's enhanced reasoning capabilities enabling more reliable outcomes, encouraging delivery-focused promises. So you can see it isn’t fool-proof but everyone seems to know this isn’t going anywhere.
So…you have some people going all in, others keenly aware but choosing to not act until further legislation and the obvious normie plugged into the matrix blissfully unaware…
Which one are you?
Applying the Series Guidance for Consistent Delivery
For arrangements with recurring GenAI outcomes, ASC 606's series guidance allows bundling substantially similar services into one performance obligation if they transfer similarly. This applies when AI resolutions are uniform and provided over time.
Key Criteria:
Substantially the Same: Each outcome (e.g., query resolution) must mirror others in nature.
Same Transfer Pattern: Delivery occurs progressively, often measured by output (e.g., resolutions completed).
If met, revenue can be recognized as a single stream, simplifying accounting. However, 2025 trends show scalability hurdles: As AI volumes spike—driven by events like the AI market's 2024 revenue of $279 billion—ensuring uniformity becomes tough amid evolving models.
AI is evolving faster than any legislation can be proposed.
Revenue Recognition Patterns: Five Essential Considerations
Recognition varies by promise type and fee structure. Here's a streamlined overview:
In practice, outcome-based models thrive in stable environments but falter in volatile ones.
For instance, the Futurum Group's 2024 forecast predicted traction for outcomes pricing in AI agents by 2025, yet past failures—due to unmeasurable results—highlight risks.
Zuora's 2025 insights reveal many SaaS companies missing monetization opportunities, as rapid AI evolution outpaces pricing strategies. What a shame!
Recent World Events: Boosters and Barriers
Supporting Factors:
Market Boom: Explosive growth in AI SaaS startups (60+ notable in 2025) and trends like deep learning for automation validate outcome pricing's potential.
Industry Shifts: Events like CES 2025 showcased AI-SaaS integrations, with vertical specialization and usage-based hybrids paving the way for outcomes.
Efficiency Gains: Reports from TTMS and RevenueGrid emphasize hyperautomation and human-AI collaboration, enabling reliable outcomes.
Detracting Factors:
Implementation Hurdles: Criticisms abound, including outcome confusion (outputs vs. true business results) and solution immaturity, as per BCG and Monetizely.
Economic and Ethical Pressures: The 2024–2025 global tech slowdown, coupled with energy crises and ethical debates (e.g., PwC on AI governance), could deter investments.
Regulatory Waves: The EU AI Act's 2025 enforcement phases introduce compliance costs, potentially complicating cross-border outcome metrics.
Moving Forward: Strategies for Success
This has rightfully led many to beg the question: is the current SaaS model (pay for access per month [generally] going to be rendered valueless, $0, because the internet is turning into mainly AI agents doing the work of current pay-to-access software…completely for free? Making even the current best paid for software almost…silly?
Valid question.
GenAI's integration into SaaS heralds a value-centric era, but outcome-based pricing demands nuanced revenue handling (not yet perfected). With global AI investments soaring yet tempered by practical and ethical challenges, proactive adaptation will separate leaders from laggards.
In a way, it’s true, in another…it isn’t. Depends entirely on how you, as the owner and creator of your saas, channel or wallet, respond with this information.
What is really is…is: an opportunity to sieze if…you are a builder.
Build with these things in mind and you will benefit!
What do you think? Leave a comment.
God-Willing, see you at the next letter.
GRACE & PEACE