AI has basically made every B2B and SaaS product at least 10%-20% better now. In some cases, 50%-100% better. Everyone has rolled out something pretty cool by now, that really works. The problem, and I hate this term, but “AI Natives” are still … just better at it. The LLMs are open to everyone. It’s just, if you started there, and it’s all you do, you still are in many cases … better at it. So incumbents being 10%-50% better than 18 months ago still isn’t enough to reignite growth, in many cases. It may forestall churn, and lean to some upsell, but even 50% “better” with AI may not be enough to grow. You have to actually catch the startups, for real, maybe in many ways, for the first time ever in B2B.

"How to become hyper-employable today in B2B? Deploy 1-2 agents for real, yourself. To your GTM team, to your company. And deliver results. Don't hire someone to do it. Don't have an agency do it. Do it yourself. You'll learn so, so much."

New!! The SaaStr AI Guide to Building a Top-Tier Inbound AI Agent 17,000 conversations. ~600 meetings booked. 60% more new business. 3 humans behind it all. The old way was a form letter a day later. Almost anything beats that.

So almost every pre-AI B2B product we use has told us they are raising prices for agent access. Salesforce and HubSpot are two of the latest, albeit differently (the price increases at HubSpot seem to be mainly for their own agents, while Saleforce is increasing prices for third party agents). But it’s not just them. Another niche CRM we’ve used for 5+ years just told us we have to pay extra for agent use of API. Another niche product we use just deprecated the API we were using for 10K. Maybe the only apps not doing this … are the ones that started agentic. What I worry is the vendors aren’t thinking this through. The immediate reaction of our agents to agentic price increases is … How Can We Work Around Them. Not cheat. But literally to move data off all these platforms so there is no API tax for the agents to use them. And the humans naturally will want to stop having their agents use apps that raise steep prices to use them. So I do worry these vendors are inadvertently creating an agentic death spiral. Usage will drop, new agents will move off the platforms as much as practical, and these API tools will lead to less usage, less stickiness, and … less of a moat.

"By the end of 2026, agents will read more Stripe docs than humans. Your job is to sell to agents in 2026. You have to be the vendor of choice not just for humans, but for agents." with @MaiaJo_ AI CRO @stripe

I've done a lot of podcasts with @HarryStebbings since 2013 (!). Well before the current show we do with Rory. Without question he makes it easier than anyone else to ... do it. No endless emails or EAs, no drama, no agencies. Just removes all friction. Removing friction always works. Sales, CX, marketing, everything.

Interesting. I can only assume that like Twilio, Bird has massively benefitted from agentic and AI spend Twilio seemed left for dead just 18 months ago, with growth having dropped to single digits. Now it is growing 22%+ and accelerating. Bird (a competitor) is seeing the same.

"If a sales rep can't provide value in the Age of AI, customers won't buy. And that means know the product cold. And agents of course will just bypass them. You just have to be a true expert in the product you sell in the Age of AI."

New!! Being a System of Record Helps With Retention. But Alone, It Won't Equal Growth ServiceTitan cut off Podium, kept all 1,000 customers. That is retention. Growth is Snowflake +34%, Databricks $7B. The record is the floor, not the engine.

Dreamforce last week was an epic rebirth in many ways of AI for CRM (and more). I'll share more thoughts soon. One memory of Dreamforces Past I had the other day was when we took one of our largest then-customers (Groupon) out to a fancy dinner after Dreamforce 201X (not sure which year). At the time, they were a Top 5 customer for us, with 2,000+ sales reps using EchoSign (now AdobeSign) for Salesforce. Our lead at dinner turned to me and said, "This is a pretty fancy dinner. That probably means I'm overpaying." It sounds like a joke, but it wasn't, and it did make me reflect. You always want customers to feel like they are getting good value, no matter what they pay in absolute terms. And they can always smell any signs they aren't. Mediocre support, upcharges that shouldn't be there, a CS team that is only there to upsell, an agent that is too hard to train, etc. P.S. They were getting a great deal!

"Look we've built 12+ AI Agents ourselves at SaaStr, including an AI VP of Revenue. They're great. But man ... buy don't build if you can. It's so much work to maintain them. We only build what we can't buy." with @lennysan

Has some blockers tonight and switched to Astra and they were gonna Will never know for sure if that's why but that's my report for today Also got Jev working for real it's great if you don't ask too much of it and carefully tune prompts

New!! The New Career Path: Joining a 0% Grower. Maybe For Many, It's Better. A flat $150M ARR company at 88% retention must close $18M of new ARR a year. Growing 0% is real work. The classic playbook works on day 1. For many execs, that's better.

An AI agent set 682 meetings. Not a human. Jason Lemkin: "682 strong meetings. That is better than any human BDR I've ever worked with." If you do nothing else, automate inbound.