Jason’s Takes on This Week’s 20VC: Loyalty Changed for Good, Agents Now Pick Your Vendors, and When to Take the $2 Billion

The Shift From Humans to Silicon Gatekeepers

I’ve spent most of my career watching the B2B sales cycle evolve, but we are entering a phase that feels fundamentally different. The old guard of SaaS relied on the concept of 'the relationship.' You took a VP of Engineering to lunch, you built trust, and you signed a three-year contract. That world is dissolving. As we look at the insights coming out of the latest industry discussions between veterans like Dev Ittycheria and Harry Stebbings, a clear pattern emerges: the agents are taking over.

When we talk about AI agents picking your vendors, we aren't just talking about a better search engine. We are talking about a fundamental shift in procurement. For a founder, this is terrifying and liberating at the same time. If a piece of software is deciding which database or CRM to use based on performance metrics and API compatibility rather than who has the best golf game, the meritocracy finally arrives. But if you're still selling like it's 2015, you're already obsolete.

Loyalty is a Depreciating Asset

One of the hardest pills to swallow in the current market is that customer loyalty has changed for good. In the past, switching costs were the moat. If it took six months to migrate data, you stayed with your vendor even if they raised prices or let their product stagnate. Today, the friction of moving between services is reaching an all-time low. AI-driven migration tools and standardized data layers mean that the 'lock-in' strategy is failing.

Builders need to understand that loyalty now has to be earned every single month. The moment a competitor offers a 10% efficiency gain that an AI agent can detect, your customer might be gone before your account manager even realizes there was a problem. We have to stop building for retention via friction and start building for retention via continuous value. If your product doesn't stand up to a cold, hard logic test performed by an algorithm, you have no moat.

The Two Billion Dollar Dilemma

There is a lot of talk about the 'right time' to sell. The $2 billion mark is often cited as a psychological and financial threshold for founders. But the recent discourse highlights a more nuanced reality. Taking the money isn't just about the exit; it’s about the lifecycle of the problem you are solving. If you are building in a space where AI is rapidly commoditizing your core value proposition, that $2 billion offer looks a lot better today than it might in eighteen months.

I see too many founders holding out for a 'generational company' status when they are actually building a feature that will be integrated into a larger ecosystem within two years. Honesty is the rarest commodity in the founder's office. You have to ask yourself if you are truly building the foundation of the next decade, or if you are simply optimizing a workflow that an LLM will soon handle natively. If it's the latter, you take the win and move on to the next problem.

Marketing to the Machine

If agents are picking vendors, how do you market? Your documentation is your new sales deck. Your API latency is your new brand identity. We are moving toward a 'headless' sales process where the primary audience for your technical specifications isn't a human developer, but an LLM-based agent tasked with optimizing a stack.

This means builders need to stop over-investing in flashy landing pages and start investing in machine-readable transparency. If an agent can't scrape your pricing, your limitations, and your integration capabilities, you won't even make the shortlist. The 'Contact Us for Pricing' era is dying because an agent won't wait for a discovery call. It will simply move to the competitor with a public pricing tier and a robust documentation library.

The Founder's New Mandate

What does this mean for you, the person actually writing the code and leading the team? It means you have to be more skeptical of your own success. The metrics that used to signal health—like high NPS or low churn—might be lagging indicators that don't reflect the new reality of AI-driven procurement.

• Audit your friction: If it’s hard to leave your product, you aren’t safe; you’re just a target for a migration startup.
• Optimize for agents: Make sure your technical specs are as clear to a bot as they are to a human.
• Re-evaluate your exit: In a world where AI moves this fast, the window for massive exits can slam shut faster than we've ever seen before.

The transition from human-led sales to agent-led selection isn't a future trend; it's happening in the background of every procurement department right now. The founders who survive this shift will be the ones who stop trying to win hearts and start trying to win the logic gate. It’s not personal, it’s just efficient. And in this market, efficiency is the only thing that lasts.

Read the original at SaaStr →