Product-market fit:
How breakout startups unlock and sustain demand
Business leaders reflect on what it takes to create products customers can’t do without
Netradyne, the AI-powered fleet safety company, learned a similar truth: Lasting change starts with recognition, not reprimand. While competitors focused solely on negative events, Netradyne built its platform around positive reinforcement, celebrating and reinforcing safe driving as part of its coaching, rather than simply flagging mistakes. The result was deep emotional buy-in from truck drivers themselves, creating loyalty and advocacy that naturally flowed to the fleet managers who invested in the system.
“Everyone would rather be praised than punished,” says Adam Kahn, Netradyne’s chief business development officer. “When feedback starts with what’s right and builds on it, trust replaces tension — from ‘you’re watching me’ to ‘you’re with me.’”
The caveat: Focusing on happy customers only works when their enthusiasm aligns around common themes that can scale. If every user loves something different, there may not be enough signal to build on. Asking customers “What type of people would most benefit from our product?” can help clarify whether there’s a broader market behind your early adopters.
Manu Sharma, co-founder and CEO of Labelbox, recalls going on an extensive listening tour, meeting enterprise customers and major tech companies to understand where the market was heading. The conversations yielded clear insights: AI leaders like OpenAI and Google urgently needed frontier data and model evaluation capabilities to fuel their latest breakthroughs. Labelbox’s pivot was a bold one. The company shifted from selling AI tools to enterprises to using those same tools to generate the critical training data for the labs and disruptors building the next generation of AI. “It wasn’t easy,” Sharma says. “We had to take a major risk, rebuild parts of the company, and rethink our operating model from the ground up.”
The bet paid off: Labelbox is now a trusted partner for the AI labs shaping the future of the industry, with a network of 1 million domain experts ensuring the data behind cutting-edge models is accurate, reliable, and ready for deployment. “In an industry this dynamic, we have to stay grounded in the real value our customers get every day, while also anticipating where the puck is going,” Sharma says.
The lesson: PMF can be lost and won again, but only when companies are willing to listen deeply and adapt quickly.
Even industry giants have had to reinvent themselves to maintain PMF. Instagram, for instance, started as a mobile check-in app, similar to Foursquare, before noticing that users were drawn far more to its photo-sharing feature than the check-ins or gaming elements. When it leaned into photo-sharing, it found success on a scale that dwarfed its early traction.
