GaaS Is Coming: What Jensen Huang’s GTC Keynote Means for Growth Teams
Learn why seat compression is killing SaaS revenue, how Intercom's $0.99/resolution model hit nine-figure ARR, and 5 moves growth
Jensen Huang stood on stage at GTC 2026 last week and said something that should terrify every SaaS growth team still optimizing for seat expansion.
“Every SaaS company will become an agentic-as-a-service company, and every engineer will carry an annual token budget alongside their salary.”
That’s a direct quote. And if you’re running growth at a SaaS company, it changes everything about how you measure, price, and scale your product.
The SaaS model we all know, charge per seat, expand by adding users, measure growth by net seat expansion, is cracking. Salesforce stock is down 26% this year. Adobe is at multi-year lows. Both for the same reason: customer seat compression. Their customers are using AI to do more with fewer people, which means fewer seats, which means less revenue under the old model.
Meanwhile, Intercom’s Fin AI agent hit nine-figure ARR by charging $0.99 per resolution. Not per seat. Per outcome.
This is the thesis: SaaS is shifting from selling access to selling outcomes. Companies that get this right will eat the ones that don’t. And growth teams are the ones who need to lead the transition, because this rewrites every metric you care about.
Here’s how this changes everything.
The Seat-Based Model Is Dying (and Your Growth Metrics Are Dying With It)
Seat compression is already here
Salesforce lost 26% of its stock value since January 2026. The culprit? Customers are replacing human operators with AI agents. One customer who used to need 50 Salesforce seats now needs 15. Salesforce’s own Agentforce product is growing at 114% YoY, but it’s actively cannibalizing their seat-based revenue faster than it can replace it.
The irony is brutal: Salesforce built AI that helps customers need less Salesforce.
Your expansion metrics are lying to you
If your North Star is “net seat expansion,” you’re measuring a dying signal. AI agents don’t buy seats. They consume outcomes. A company deploying Intercom Fin doesn’t add more support agents to the plan. They resolve more tickets, and Intercom bills them $0.99 per resolution. The “expansion” isn’t horizontal (more users). It’s vertical (more value delivered per account).
The inference economics tipping point
Jensen Huang announced the Vera Rubin chip at GTC 2026, projecting a 10x reduction in inference token costs. Tasks that cost $10 per million tokens today become $1 per million tokens. That makes AI agents economically viable for use cases that were too expensive last year. More viable AI agents = more seat replacement = faster seat compression.
Gartner projects 40% of enterprise SaaS contracts will include outcome-based pricing components by end of 2026. That’s not a prediction for 2030. That’s this year.
The New Model: From Seats to Outcomes (GaaS in Practice)
The companies winning right now aren’t bolting AI onto seat-based models. They’re rebuilding around a completely different unit of value. Here’s the framework.
Pillar 1: The outcome is the product
Intercom doesn’t sell “AI-powered support software.” It sells resolved support tickets. The product IS the outcome.
Intercom Fin has processed 40M+ resolved conversations with a 67% resolution rate. At $0.99 each, do the math: that’s a nine-figure ARR product growing at 393% annualized. Zendesk followed with $1.50-$2.00 per automated resolution.
The pattern: define a measurable outcome, price it directly, then scale the outcome volume.
Pillar 2: The metric shifts from users to throughput
In SaaS, growth means more users. In GaaS, growth means more outcomes processed. Your dashboard changes:
Old: Monthly Active Users → New: Monthly Outcomes Delivered
Old: Seats per account → New: Outcome volume per account
Old: Net Revenue Retention (driven by seat expansion) → New: Net Revenue Retention (driven by outcome volume growth)
Old: CAC per user → New: CAC per outcome-generating account
This isn’t a cosmetic rename. When your unit of value changes, every downstream metric, every growth loop, every pricing page, every sales motion has to change with it.
Pillar 3: Expansion becomes vertical, not horizontal
In seat-based SaaS, expansion = getting more people to use the product. In outcome-based GaaS, expansion = getting each account to process more outcomes through your system.
This flips the entire expansion playbook. Instead of “land one team, expand to adjacent teams,” the motion becomes “land one use case, expand the outcome volume within that use case.” Then expand to adjacent outcome types.
Pillar 4: Pricing becomes the growth lever
In SaaS, pricing is a finance decision. In GaaS, pricing IS the growth strategy.
Intercom’s $0.99/resolution isn’t just a price point. It’s a growth mechanic. It removes the buyer’s risk (”I only pay when it works”), which accelerates adoption. It creates a natural expansion path (more tickets = more revenue without a sales call). And it aligns incentives: Intercom only makes money when the customer gets value.
Chargebee’s 2026 AI pricing guide calls this the “outcome-as-acquisition” model. The pricing itself is the growth loop.
Pillar 5: The moat moves from switching costs to outcome quality
Per-seat SaaS has a natural moat: switching costs. Moving 500 people off Salesforce is painful, so customers stay even when unhappy.
Outcome-based pricing weakens this moat. If I’m paying per resolution and a competitor resolves tickets better at a lower cost, switching is just redirecting my ticket flow. No user retraining. No data migration (in many cases). Just a better outcome engine.
The new moat: outcome quality. How well do your AI agents perform? What’s your resolution rate? Your accuracy? Your time-to-resolution? Growth teams need to optimize for outcome quality metrics the same way they used to optimize for activation and retention.
Why Most SaaS Companies Will Botch This
Here’s my contrarian take: the majority of SaaS companies are going to fail at this transition, and it won’t be because of technology. It will be because of incentives.
Most SaaS companies will bolt AI onto their seat-based models and call it innovation. That’s like putting a jet engine on a horse carriage and calling it a plane.
The bolt-on trap
Salesforce’s Agentforce is the poster child. It’s growing 114% YoY. It’s genuinely useful technology. And it’s destroying Salesforce’s core business model. Salesforce stock dropped 30% YTD despite the AI product succeeding. Why? Because the company hasn’t rearchitected its revenue model around outcomes. The AI is an add-on to a seat-based structure.
This is the trap most companies will fall into: shipping AI features while keeping the per-seat pricing model intact. The result? The AI makes each seat more productive, the customer needs fewer seats, and revenue contracts.
The org chart problem
Transitioning from seats to outcomes requires growth teams and finance teams to agree on a completely different set of KPIs. That’s a multi-quarter internal battle most companies aren’t ready for. The VP of Sales whose comp is tied to seat expansion will fight outcome-based pricing. The CFO who models revenue on predictable per-seat contracts will resist variable outcome-based revenue.
The companies that will win are the ones where the CEO forces this from the top down. Intercom did it. Des Traynor bet the company on outcome-based pricing for Fin. Most CEOs won’t have that conviction.
What I’m Seeing in Practice
I work in growth at an AI video platform (ngram). The conversations we’re having internally mirror exactly what I’m describing here. When your product uses AI to generate videos, the question quickly becomes: do you charge per seat, per video generated, or per business outcome achieved?
The answer changes your entire growth model. Charge per seat, and you’re incentivized to get more people logging in. Charge per video, and you’re incentivized to make video creation addictive. Charge per outcome (views, engagement, leads generated from videos), and you’re incentivized to make the videos actually work for the customer.
Each model creates different growth loops, different expansion motions, and different retention dynamics. The companies figuring this out now will have a massive head start.
I’ve also been watching how growth loops compound differently when the unit of value shifts.
In seat-based SaaS, the growth loop is:
user invites colleague → new seat → more revenue.
In outcome-based GaaS, the growth loop is:
agent delivers outcome → customer trusts system more → sends more volume → more revenue.
The second loop compounds faster because it doesn’t require a human decision at each step.
How Growth Teams Should Prepare (5 Moves to Make This Quarter)
1. Audit your “unit of value” today. What does your customer actually pay for? What do they actually VALUE? If those are different things, you have a GaaS opportunity. Map every feature to the outcome it drives, and identify which outcomes customers would pay for directly.
2. Build outcome tracking into your product analytics. If you can’t measure the outcome, you can’t price it. Start instrumenting outcome events (resolutions, completions, generations, qualifications) alongside your existing user behavior analytics. You need this data before the pricing conversation even starts.
3. Model what happens to your revenue when seats compress 30%. Because they will. Run the scenario: if your average customer needs 30% fewer seats in 18 months (a conservative estimate), what happens to your ARR? Now model the same scenario with outcome-based pricing layered in. Present both models to your leadership team.
4. Experiment with hybrid pricing now. You don’t have to go full outcome-based overnight. Test a hybrid: base platform fee + outcome-based variable component. Intercom did this. They still have a platform subscription for core features, but Fin’s pricing is purely outcome-based. Start small, measure, and expand.
5. Redefine your North Star metric. If it’s still “seats” or “MAU,” it’s time to evolve. What’s the outcome metric that best represents the value your product delivers? For Intercom, it’s resolutions. For an AI video tool, it might be videos that drive measurable engagement. For a sales tool, it might be qualified meetings booked. Find yours.
The Growth Teams That Move First Will Own This Decade
Jensen Huang projected $1 trillion in AI-driven revenue from 2025-2027. That money isn’t going to companies optimizing for seat expansion. It’s going to companies that figured out how to sell outcomes, price them correctly, and build growth loops around outcome volume.
The SaaS-to-GaaS transition will be as disruptive as the on-premise-to-cloud transition was. And just like that shift, the companies that move early will define the market, and the ones that move late will spend years playing catch-up.
My prediction: by 2028, “per-seat pricing” will feel as outdated as “per-server licensing” feels today.
If you’re in growth, this is your moment. The playbook is being rewritten. Be the one holding the pen.
If this shifted how you’re thinking about pricing and growth metrics, share it with your team. This is a conversation every SaaS growth org should be having right now.
FAQs
What is GaaS (Growth-as-a-Service)?
GaaS refers to the shift from traditional SaaS seat-based pricing to outcome-based pricing models. Instead of charging per user, GaaS companies charge per measurable outcome delivered — like Intercom charging $0.99 per support ticket resolved by their AI agent Fin.
Why is per-seat SaaS pricing declining?
AI agents are replacing human operators, causing seat compression. Companies that once needed 50 software seats now need 15. Salesforce lost 26% of its stock value in early 2026 partly because its own AI product, Agentforce, helps customers need fewer seats — cannibalizing its core revenue model.
What is outcome-based pricing in SaaS?
Outcome-based pricing charges customers for measurable results rather than access or users. Examples include Intercom Fin at $0.99 per resolution and Zendesk AI at $1.50–$2.00 per automated resolution. Gartner projects 40% of enterprise SaaS contracts will include outcome-based pricing components by end of 2026.
How does the SaaS-to-GaaS transition affect growth metrics?
Traditional metrics like Monthly Active Users and net seat expansion become unreliable. Growth teams need to shift to outcome-volume metrics: Monthly Outcomes Delivered, outcome volume per account, and CAC per outcome-generating account. Expansion becomes vertical (more outcomes per account) rather than horizontal (more seats).
How should growth teams prepare for outcome-based pricing?
Start by auditing your unit of value, building outcome tracking into product analytics, and modeling what happens when seats compress 30%. Experiment with hybrid pricing (base fee + outcome-based variable), and redefine your North Star metric around the outcome your product delivers rather than user counts.
What did Jensen Huang say about SaaS at GTC 2026?
Jensen Huang stated that “every SaaS company will become an agentic-as-a-service company” and projected $1 trillion in AI-driven revenue from 2025–2027. He also announced the Vera Rubin chip, which projects a 10x reduction in inference token costs — making AI agents economically viable for far more use cases.