The Numbers That Matter
SaaStr AI 2026 in San Mateo delivered the clearest picture yet of what AI-augmented sales actually looks like in production. Not theory. Not pilot programs. Real GTM orgs with real quota pressure shipping real results.
Anthropic: Eleanor Dorfman, Head of Industries, rebuilt the enterprise sales motion after a Claude release sent demand vertical. Four months later, 54% of new enterprise logos close self-serve. Real enterprise contracts, real ACV, no rep gating the door. The reps who used to run those deals now focus on accounts where a human changes the outcome.
Owner.com: Kyle Norton, CRO, put up the most concrete rep economics of the event. His AI-native team averages $2M+ ARR per rep per year. That is 20x their OTE, not 3x. Outbound BDRs are closing $100K+ per month in actual revenue, not pipeline. The comp benchmark just moved.
Gamma: Grant Lee's team hit $100M ARR with 50 employees, 50 million users, and 600,000 paying subscribers. Almost no sales team. His biggest regret: waiting too long to add one. If the company with the strongest excuse to skip sales wishes it had moved sooner, most founders riding inbound are later than they think.
PayPal/Salesforce: Adam Alfano (President, Salesforce) and Eitan Saban (Head of North America Mid Market Sales, PayPal) pointed Agentforce at 8,000 leads per month that no human was going to touch. The accounts sitting at the bottom of the funnel that used to rot. The agents are working them.
What Changed
The session lineup included Vercel (COO Jeanne DeWitt Grosser on a lead agent that took a 10-person function down to 1), Replit (data showing rep-level AI usage predicts quota attainment), Stripe (GM of Enterprise Product Maia Josebachvili on the four patterns behind the fastest-growing AI companies), and Monaco (Co-Founder and CEO Sam Blond on comp, headcount, and margin math when agents deliver the outcome).
Nobody on stage was debating whether to put agents in the revenue org. They already did it. The war stories were about what broke, what worked, and what they would do differently next time.
The Tension
Gamma says most companies add sales too late. Anthropic says when demand spikes, the instinct to throw reps at it is often the wrong first move. Both are right. The way to hold both: treat human selling as expensive and scarce, spend it only where it moves the deal, and let everyone else buy without waiting on a calendar.
Owner.com is running reps at $2M ARR each in the SMB segment, selling $10K ACV software to independent restaurants. If your reps are running at 3x or 4x their comp and you think that is healthy, the ceiling just moved. AI-augmented reps at a well-run org are producing at a level that makes the old benchmarks look outdated.
What It Means for ANZ
The playbooks are shipping. Anthropic is hiring into Industries, which means vertical/industry GTM rather than only horizontal enterprise. Salesforce is productising agentic sales with Agentforce. Anthropic's Claude is integrating into Salesforce. The enterprise sales stack is rebuilding around agents, and the early movers are already posting numbers that reset the benchmarks.
For ANZ sales teams: the question is not whether to ship AI, it is how far behind you are willing to fall while you debate it. The comp math, the headcount models, and the quota expectations are all moving. The teams that figure this out first will reset the market for everyone else.