For years, the prevailing wisdom in tech circles was simple: build a great product and customers will come. Product-led growth became gospel, especially among AI-native companies. Sales teams? Those were relics of enterprise software's past—expensive, slow, and unnecessary when your product could sell itself.
Replit's founder and CEO Amjad Masad was a vocal skeptic of traditional sales culture. For nearly a decade, Replit grew almost entirely through product-led growth, with Masad's Twitter account serving as the company's entire marketing function. No sales team. No account executives. Just a single person handling inbound inquiries alongside three other jobs.
Then reality arrived. Companies whose employees were already using Replit started calling, asking how to buy it properly. The lone rep was overwhelmed. Customer dinners and direct engagement proved effective. And Masad made a remarkable public admission: he now expects more than half of Replit's workforce to be salespeople by year-end.
The Pattern Across AI-First Companies
Replit isn't alone. This pattern—delaying but eventually building a sales team—now extends across the most successful AI-native companies. Gamma, Lovable, Anthropic, and Replit all reached $100 million to $250 million in ARR before hiring significant sales capacity. That's later than the pre-AI era threshold of $20 million to $50 million, but the destination is the same: everyone ends up with a sales team.
The math is straightforward. Freemium and self-serve can carry you further now than five years ago. PLG works brilliantly for individual users and small teams. But when enterprise demand arrives—when Fortune 500 companies want procurement processes, multi-year contracts, and dedicated support—automation hits a wall. You need humans to close those deals, expand accounts, and navigate organizational complexity.
What This Means for Founders
The lesson isn't that sales teams are bad or that product-led growth doesn't work. It's that they're sequential, not mutually exclusive. A working, sellable product buys you time. Early traction through PLG validates your market and generates demand. But that time is finite.
When inbound inquiries start overwhelming your team, when enterprise customers are asking for sales conversations, when revenue expansion requires account management—that's your signal. Waiting too long means leaving money on the table and frustrating ready-to-buy customers. Moving too early burns cash on salespeople who have nothing proven to sell.
The key is having a product worth selling in the first place. Not a prototype. Not a demo that breaks under real usage. A working, sellable product that solves real problems and can handle enterprise-scale deployment.
Key Takeaways:
- AI-first companies like Replit, Gamma, and Anthropic all eventually built sales teams, just later than traditional SaaS
- Product-led growth now carries companies to $100M+ ARR before sales teams are essential, versus $20-50M previously
- Enterprise customers demand human interaction for procurement, contracts, and complex deployments
- Sales and marketing typically become the largest expense line in B2B software as you scale
- The trigger point is clear: when inbound demand outpaces what automation can handle, hire sales talent
- A working, sellable product is the prerequisite—sales teams amplify what already works
Build Something Worth Selling
Here's the uncomfortable truth: if your product isn't working reliably, no sales team will save you. If it's still a prototype that breaks under real usage, enterprise customers will churn. Sales amplifies product-market fit; it doesn't create it.
That's why the sequence matters. Build a working product first. Get early traction through PLG. Validate that people will pay. Then, when enterprise demand arrives and you're turning away revenue, build the team to capture it.
The AI era hasn't eliminated sales teams. It's just shifted when you need them. Plan for that transition now, even if it's eighteen months away. Recognize that sales will eventually be your largest expense. And above all, make sure you have a product that's actually ready to sell at scale.