From $500K Stuck to $10M ARR: Vapi Founder Jordan Dearsley

Jordan Dearsley burned down a profitable business, moved to San Francisco, and stumbled into voice AI by building himself an AI therapist for walks. A year later, Vapi was at $10M ARR. His story is the clearest blueprint for what product-market fit actually feels like — and what it doesn't.

Key Facts

• ARR at pivot: ~$500K over 3 years (calendar app) (Jordan Dearsley)
• Time to $10M ARR: Approximately 12 months after launching Vapi (Jordan Dearsley)
• Team size at $60K MRR: 2 people (Jordan + co-founder Nikhil) (Jordan Dearsley)
• First pricing model: ~5 cents per minute, pass model cost at cost — still the model today (Jordan Dearsley)
• Growth method: Zero paid ads; Discord community, Product Hunt launch, YouTube organic (Jordan Dearsley)

Why Jordan Walked Away From $500K ARR

Jordan didn't quit because the business failed — he quit because he stopped believing it mattered. After three years building a calendar app to $500K ARR, the honest signal was internal: he no longer had the will to keep going. That clarity was the pivot.

Many founders stay too long in the gray zone — enough revenue to feel legitimate, not enough to feel like a real company. Jordan Dearsley lived there for three years with a calendar and AI note-taking app that grew to $500K ARR but never had a grander vision behind it.

The break came from a blunt conversation with YC group partner Michael Seibel. As Jordan recalled: 'He was just like, Guys, what the fuck are you doing? You don't care? You're working 9 to 5. What are you doing with your lives?' That jolt made Jordan realize he'd been lying to himself about the company's potential.

His advice to founders in the same position: treat your existing customers as an asset, not a sunk cost. 'Talk to those people and find out what else you could do in their daily lives. What are the top five pains? You're at five — why didn't you get me solving one?' The revenue you've earned buys you access to real problems, if you're willing to ask.

"I did not have it in me to put time into this and feel like my time was worthwhile anymore. I had convinced myself for long enough that it was worth my time, but I think it just finally became clear to me that I was lying to myself." — Jordan Dearsley

"We got stuck in a rut of chasing a local maxima. Moving was like the shock to the system that really did it." — Jordan Dearsley

• Revenue without conviction is a trap — not a foundation.
• Seibel's challenge: 'You don't care. What are you doing with your lives?'
• Moving cities was the 'shock to the system' that forced a real reset.
• Three months of darkness followed — including visits to food banks looking for problems to solve.

How Vapi Was Born From an AI Therapist Walk App

Vapi didn't start as a voice AI platform. It started as Jordan's personal AI therapist he could call on walks. When he realized the infrastructure to support fluid AI voice conversations didn't exist, he built it — and that infrastructure became the product.

Frustrated with chat-based AI tools, Jordan built a voice companion he could call during walks to think through startup problems. 'I don't want to fucking like chat with this thing. I hate typing. I just want to walk and think and have a fluid conversation with something,' he said. When he searched for an existing solution, he found nothing.

Building it forced him to wrangle with transcription models, LLMs, and text-to-speech APIs — none of which were designed for real-time voice. He discovered a YC startup called Hyperbound struggling with the exact same orchestration problem. He offered them his stack, charged five cents a minute, and had his first customer.

That pricing model — usage-based, passing model costs at cost — is still Vapi's model today. The first use case was AI sales role-play training. Hyperbound eventually closed a deal with a real sales team, validating that the underlying infrastructure had commercial legs.

"I looked around and it was like, wow, it literally exists nowhere. There is nothing — at least pre-4.0 audio — that allows you to have a fluid conversation." — Jordan Dearsley

"Why don't you just use my thing? I'll rip off the therapy thing. I'll give you an endpoint to hit, and it'll set up a WebSocket connection, and audio will stream in and out." — Jordan Dearsley

Finding the 10/10 Burning Pain: Vapi's Product-Market Fit Signal

Product-market fit at Vapi showed up as customer rage during outages. Even with just eight users, every one of them was furious when calls dropped — because their businesses depended on it. That existential anger, not revenue, was Jordan's clearest signal.

Jordan's framework for validating a real problem is deceptively simple: ask the customer to rate their pain on a scale of 1 to 10, then push back until they prove it. 'Ask them, on a scale of 1 to 10, how painful is this? You'll hear six, and you're like — wait, what? It's very important,' he said.

The signal that separated Vapi from the calendar app wasn't growth rate — it was the quality of customer anger. With the old product, outages went unnoticed. With Vapi, even eight users would escalate immediately because their own customers were affected. 'Usually if you have an outage and you have eight users, nobody notices,' Jordan noted. 'Where it's like — oh no, their customers are pissed at them because they didn't get their calls answered. And so they lost money.'

His definition of product-market fit is deliberately unglamorous: 'Probably after we were at like a couple million in revenue and we had an outage and everyone was just so fucking mad — that was probably it.' The lesson: stop waiting for a spiritual moment. Look for the users who can't afford for you to fail.

"Find someone who has a 10 out of 10 burning pain. When I say 10 out of 10, I mean literally ask them, on a scale of 1 to 10, how painful is this? I know it sounds like a silly question, but just ask them." — Jordan Dearsley

"Don't think it's a really big thing. Product market fit can be achieved with one person and one user. You're going to have product market fit immediately with an extremely small market as N of one. Then you just need to achieve it with two, and then four, and then five." — Jordan Dearsley

• Ask customers to rate their pain 1–10. Push back if they say anything under 9.
• Real PMF shows up as rage during outages, not satisfaction scores.
• N=1 is enough to start — then find N=2, N=4, N=10.
• Existential pain for the customer means your product is load-bearing for their business.

How Vapi Grew to $10M ARR Without Paid Marketing

Vapi grew from $200/month to $10M ARR through a Product Hunt launch, Discord community, and organic YouTube coverage — zero paid ads. The Product Hunt launch alone 3x'd call volume. From there, compounding weekly growth of 10–20% did the rest.

The growth trajectory was methodical rather than explosive at first. In January 2024, Vapi processed 32,000 calls — almost entirely from one sales training startup. By March, after a Product Hunt launch amplified by a free video from a Discord community member and retweets from well-followed contacts, call volume hit 164,000.

From there, the doubling continued: 279K calls in April, 464K in May, 692K in June, 1.1M in July. By mid-2024, Vapi was at roughly $4M ARR with a team of around six people. 'We did zero ads. We did zero marketing. We didn't have the capacity for marketing. We just had this Discord community and maybe some people make YouTube videos about us,' Jordan said.

The harder challenge was surviving the growth. Jordan's co-founder Nikhil eventually pulled him aside: 'Dude, if you don't get me a fucking infrastructure engineer, we are going to die.' At peak chaos, Vapi had 300–400 open Slack channels with customers — before closing all of them to focus, a move that paradoxically improved their reputation.

"We did zero ads. We did zero marketing. We didn't have the capacity for marketing. We just had this Discord community, and maybe some people make YouTube videos about us. That's basically it." — Jordan Dearsley

"My co-founder pulled me aside one time and was like, 'Dude, if you don't get me a fucking infrastructure engineer, we are going to die.' And I was like, oh, I need to stop coding now. Got it." — Jordan Dearsley

Competing Against OpenAI: Vapi's Defensibility Strategy

Vapi's defensibility isn't the AI models — it's the accumulated expertise of deploying voice agents to production at scale. Every API parameter was born from a real developer pain. That institutional knowledge, and the complexity of enterprise last-mile deployment, is what Jordan believes protects Vapi from being displaced.

When OpenAI announced GPT-4o Audio in April 2024, Vapi's team had a moment of genuine fear. Their original vision — to become the speech-to-speech model company — was suddenly obsolete. A brief conversation with an investor nearly ended the company. They kept going, and GPT-4o turned out to grow awareness of voice AI broadly rather than displace Vapi specifically.

Sales advisor Mitch Miranda reframed the opportunity: enterprises can't just use OpenAI directly. The complexity of tuning, compliance, telephony integration, and production reliability requires a dedicated partner. 'That expertise is very hard to learn without having done it yourself many times,' Jordan said.

Jordan's mental model for evaluating AI startup defensibility maps directly to how he thinks about OpenAI releases: 'If them launching a new model makes your product better and you deliver more value, that's good. If it makes you scared and withers how much value you're actually creating — that's not a great place to be.' For Vapi, better models mean a better platform. That's the right relationship to have with your infrastructure providers.

"Every single parameter in our API is from us hearing from a developer, figuring out the core pain that they have, and shipping it. Our product is a representation of everything that we have learned so far." — Jordan Dearsley

"The only place that I can be safe is covering that last mile — because that's the trickiest part for a super, super large company to tackle." — Jordan Dearsley

Calendar App vs. Vapi: Two Startups, Two Outcomes