247.ai

PV Cannon's journey began in the 1990s when he developed chat as a customer service vehicle during the dot-com era. But what really sparked 247.ai came from visiting contact centers across the US. "I was just blown away by the number of people servicing the US economy," he recalls. He noticed something crucial: despite the rise of digital channels, contact centers were still treating chat like phone calls, following the same scripts as the 800-number system designed for the 1980s. "The whole paradigm of 'how can I help you' is kind of almost obsolete," PV realized. The new paradigm should be: you should already know what customers are doing, and with AI in the backend in real time, predict what they're attempting to do and be supportive of it.

Launched in 2000, 247.ai started with PV and co-founder Nags bootstrapping the initial development. In 2003, they raised their first and only major funding round with Sequoia (Mike Moritz joined the board), bringing in capital that would help establish credibility in a crowded market. PV was deliberate about raising money—not because the company needed it to survive, but because partnering with a top-tier venture firm signals to talent and enterprise customers that you belong to an exclusive community. "Whether you like it or not, the type of venture firm you partner with sets up your brand," he explains. Over 23 years, they've raised just over $20M total—a remarkable ratio for a $300M ARR company.

The company's beachhead was enterprise customers with millions of consumers struggling to handle customer inquiries across email, chat, messaging, and phone. 247.ai's sweet spot customer has over a million consumers and operates in a demanding, service-oriented industry. By focusing on high-ACV enterprise deals (averaging $250,000 per year at launch, growing to $2M annually by the time of this interview), PV built a land-and-expand playbook. The company optimized for a 12-month CAC payback period and categorized customers by potential spend: those capable of spending over $10M annually (where they'd spend aggressively on sales), versus smaller customers under $500K (where they'd be conservative). This disciplined approach meant every dollar spent on customer acquisition had a clear return timeline.

For years, 247.ai executed flawlessly—until 2018. A security breach derailed the company's momentum, essentially vanishing their entire sales pipeline. They spent about a year recovering, which is why their predicted $400M in revenue (reported by Reuters in 2017) never materialized and why they were "about the same" revenue a year before the interview. But the setback forced important realizations. PV doubled down on customer success, which had been under-resourced in earlier years. They shifted from a coverage model biased toward large customers to ensuring every single account had dedicated resources. The payoff: gross revenue churn of 8-10% and expansion revenue of 8-10%, leading to 100% net revenue retention. They also grew their team strategically—800 on the technology side, plus thousands of contractors in their services business who teach AI to improve the model. Two-thirds of revenue now comes from pure SaaS; one-third from professional services that deliver measurable outcomes and drive retention.

With $300M+ in ARR and over 150 enterprise customers, 247.ai is one of only a handful of profitable, bootstrapped (mostly) B2B SaaS companies at scale. They're generating 10% net profit margins and targeting 20% as steady state. Recently, they hired a new Chief Revenue Officer whose mandate was to stabilize metrics before aggressively chasing growth—and the latest quarter was their best in company history with 20% new revenue booking growth. When asked if he'd sell to PE firms courting him (Vista, others), PV was clear: an IPO is the goal. "We believe we can extract a lot of value and continue to serve customers better by just being standalone." That's the confidence of a founder who solved the hard problems decades ago and is now watching his thesis play out at scale.