In 1890, entire rooms of people were hired to do math. Their job title was literally “computer.”
Then machines showed up—and the job disappeared.
Now imagine something similar happening to the humble business phone call.
Every day, millions of calls go unanswered. A customer tries to book a service, schedule an appointment, or ask a quick question… and gets voicemail. For businesses, that’s not just inconvenient—it’s lost revenue.
In this episode of Ignite Podcast, Brian Bell sits down with Will Bodewes, co-founder and CEO of Phonely, a startup building AI voice agents that answer business phone calls and handle real conversations with customers.
The goal isn’t just automation. It’s conversion.
From Small-Town Wisconsin to Building AI Companies
Will’s path into startups didn’t begin in Silicon Valley—it began in a tiny town in northern Wisconsin.
So small, he jokes, that his family made up about 1% of the population.
Growing up there meant a limited set of winter activities: drinking beer, snowmobiling, or cross-country skiing. Will chose skiing—and got good enough to compete at the NCAA level.
But endurance sports weren’t the only extreme challenge he pursued. As a teenager, he set a world record skiing for 72 consecutive days around the world’s largest lake, a feat that shaped how he approaches difficulty.
That mindset—doing hard things others won’t—became the backbone of his founder philosophy.
After college, he launched his first startup experiment, building Bluetooth speakers embedded into canvas artwork. The project didn’t scale into a big company, but it did lead to a surreal moment: Chance the Rapper discovered the invention and followed him on Instagram.
Soon after, Will began a PhD in artificial intelligence. But something pulled him back toward startups.
“I always knew I was meant to build something,” he says.
The Idea Behind Phonely
Phonely focuses on a deceptively simple problem: answering phone calls.
While much of modern software revolves around chat interfaces and web forms, phone calls remain a critical part of many industries—especially service businesses like healthcare, roofing, plumbing, and insurance.
The challenge? Phones don’t scale.
One person can only talk to one caller at a time.
Will realized that if AI could handle these conversations reliably, businesses could capture far more revenue from inbound calls.
“We started working on voice agents before the technology really worked,” he explains.
It was one of those classic startup moments where the idea is obvious—but the technology is just barely catching up.
Finding Product-Market Fit the Messy Way
Phonely didn’t launch with a clear target market.
In the early days, Will sold the product to anyone willing to try it, including a first customer paying just $7 per month.
From there, the company slowly climbed the ladder:
$30/month customers
$100/month customers
$500/month customers
Enterprise deployments worth tens of thousands per month
Today, some customers pay $70,000 per month as the system scales across entire organizations.
But product-market fit didn’t arrive in a dramatic “aha moment.”
Instead, it emerged through a strange signal: customer complaints.
When users started complaining that a newly released feature wasn’t working correctly, the team realized they had built something valuable.
“If people complain that something is broken, that’s a good sign—it means they care about it,” Will says.
Why Founders Should Do White-Glove Onboarding
One of the most important lessons Will shares is surprisingly simple.
In the early days of a startup, founders should personally onboard customers.
Phonely leaned heavily into white-glove onboarding, helping clients configure their AI agents manually.
At first glance, this looks inefficient.
But it turned out to be a powerful shortcut to product insight.
When the founders were forced to set up the product themselves, they felt every piece of friction their customers experienced.
“This doesn’t work. This doesn’t make sense. Why did we design it like this?” became daily realizations.
That feedback loop dramatically improved the product.
Why Phonely Didn’t Verticalize (Even When Investors Asked)
Many investors pushed the team to build a vertical AI company—for example, voice AI specifically for insurance, healthcare, or home services.
Will resisted that pressure.
Instead, the company focused on a horizontal problem:
Helping AI agents convert phone calls into business outcomes.
Whether the call is for roofing, HVAC, legal services, or healthcare, the underlying challenge is the same:
Answer the phone
Understand the request
Qualify the lead
Schedule the appointment
By focusing on the conversion problem rather than a specific industry, Phonely built tools that work across many sectors.
The Reality of Building an AI Startup
From the outside, building an AI startup can look glamorous.
In reality, it’s brutal.
Will is blunt about the early days:
“For six months to a year, every day I thought about quitting.”
The product didn’t always work. Customers got frustrated. Competitors raised money.
But grit turned out to be more important than intelligence.
“The skill that matters most is just not giving up.”
What Most People Get Wrong About AI Right Now
One of Will’s most interesting insights challenges a common assumption about AI progress.
He believes AI capabilities are still improving—but not at the exponential pace people expect.
That means the real opportunity today isn’t inventing new models.
It’s implementing AI inside real businesses.
Most companies want AI but don’t know how to deploy it effectively. Founders who solve that gap—connecting AI to workflows, systems, and outcomes—can build enormous value.
Why Phone Calls May Be Making a Comeback
Ironically, as digital interfaces grow more complex, phone calls might become more appealing again.
Think about trying to book something online:
Log in
Complete two-factor authentication
Navigate confusing menus
Wait for an email response
Or…
Call and get the answer in 30 seconds.
The problem historically wasn’t the phone—it was the lack of available humans to answer it.
AI changes that equation.
As Will puts it, once AI removes the friction, calling may become the fastest way to get things done again.
The Bigger Mission
For Will, building Phonely isn’t just about business automation.
He sees startups as a way to create impact at scale.
The long-term goal is simple: help businesses and consumers connect more efficiently.
Better conversations lead to better outcomes—for both sides.
And if the vision works, the result could be millions of businesses with AI agents answering every call.
No more voicemail.
No more missed opportunities.
Just a smarter layer between people and the services they need.
👂🎧 Watch, listen, and follow on your favorite platform: https://tr.ee/S2ayrbx_fL
🙏 Join the conversation on your favorite social network: https://linktr.ee/theignitepodcast
Chapters:
00:01 Introduction to Will Bodewes and Phonely
00:32 Will’s Origin Story and Growing Up in Rural Wisconsin
01:01 Cross-Country Skiing and Competing at the NCAA Level
01:30 First Startup: Bluetooth Speakers Inside Artwork
02:46 From Startup Experiments to a PhD in Artificial Intelligence
03:24 The Beginning of Phonely and the Vision for Voice AI
04:03 Early Founder Lessons and the Long-Term Mindset
05:25 Why It Was the Right Time to Build Voice AI
06:25 Why Voice Is a Bigger Opportunity Than Chat
08:23 Finding Product-Market Fit for Phonely
09:36 The Reality of Building Products Customers Actually Want
10:17 Why Customers Don’t Know What They Want
11:53 The Power of White-Glove Onboarding for Startups
13:37 Early Customers and the First $7 Deal
15:19 Vertical vs Horizontal AI Platforms
17:36 AI Replacing Jobs and the Philosophy Behind Automation
19:23 Customer Stories and Impact on Businesses
20:27 Using Phonely Internally and Product QA
21:28 Surprising Insights from Customer Use Cases
23:09 Building the Voice AI Tech Stack
24:25 Phonely’s Onboarding Process and Freemium Model
27:27 Building Team Culture and Hiring Principles
28:12 Product Momentum and What’s Next for Phonely
29:13 Long-Term Vision for Voice AI
30:27 Rapid Fire: Contrarian AI Beliefs
31:58 The Most Important Skill for Founders
33:01 Hard Founder Decisions and Ignoring Investor Advice
34:40 Future Developments in AI and Browser Agents
36:00 Why Phone Calls Are Underrated
37:46 Risks in Voice AI and Regulatory Challenges
38:26 Advice for Researchers Thinking About Becoming Founders
Transcript:
Brian Bell (00:00.93): Hey everyone, welcome back to the Ignite podcast. Today we’re thrilled to have Wilbo Davis on the mic. He is the co-founder and CEO of Phonely, a company building conversational AI agents that answer business phone calls with natural voice and deep workflow integrations. Will’s background spans endurance athletics, early startup experiments, applied AI research, and now building voice systems that reshape how companies interact with customers. Thanks for coming on,
Will (00:24.428): Yeah, thanks so much. Glad to be here.
Brian Bell (00:26.624) Yeah, so I always start the episode with what’s your origin story question. We would love to hear your background.
Will (00:32.686) That’s a great question. I guess we’ll start at the beginning. So I grew up in a small town in Northern Wisconsin. I often brag that my town, my family made 1 % of my population. So it’s very, very small. I had a background of seven people, so not that big. A thousand people. Yeah. So yeah, it was small.
Brian Bell (00:48.438) I bet it was a family. Okay. Yeah, yeah. 7,000 people and you were seven of them or 700, I guess, if it’s 1%, yeah.
Will (01:01.656): I did have stoplight, so it came from a very, very small town, very rural experience, but I got into cross-country skiing, weirdly enough, because I was in northern Wisconsin, the only options are drink beer, snowmobile, and cross-country ski. So I took that third option in the wintertime. what’s that? Yeah, ice fish, which is basically drinking beer and sitting on a bucket in the ice. I never understand ice fishing, to be honest with you. So I got into cross-country skiing, got pretty good at that.
Brian Bell (01:16.96) An ice fish, right? An ice fish.
Will (01:30.606) I ended up going to college for that and competed all the way up to the NCAAs and that. After college, I went and started my first company, which didn’t really end up working out other than Chance the Rapper follows me on Instagram, which I think is a, you know, big flex right there. Yeah. First company is called Spoke Sound. was a, basically it was just like my first experience in startups. I always knew that I wanted to build something and I always just like,
Brian Bell (01:47.148) What was the first company and why would Chance follow you?
Will (01:59.704) For me, what excites me is contributing to the world. It’s like your whole entire life, you take things from the world, you use other things that people built and I wanted to give back to it. And so at the time, that was the best kind of idea that I had, which was to integrate Bluetooth speakers into Canvas photographs. So I was working on some, it’s called Distributed Mode Loudspeaker Technology. But it was cool. We worked with artists and they would put a Bluetooth speaker behind it and then they would hang it on the wall. And it was just like really kind of fun art piece that they could create.
Also worked with some hotels and stuff like that to help them hide the speakers basically And then Chance the Rapper found out about the company just through word of mouth from other artists and then like reached out followed me on Instagram and was like, yo I love your inventions. It was a crazy DM. Not gonna lie
Brian Bell (02:46.742) Wow, that’s amazing. So what happened next?
Will (02:47.822) Yeah. Yeah. So then after, after that we, or after that I, I started a PhD in artificial intelligence. had kind of like during COVID that I paused that just cause it was like, can’t, was in Australia and I couldn’t get over there. So I started working on that. Always kind of knew that I wanted to come back to startups, but I, you know, started the PhD in AI in 2020 before it was cool. And then as it continued to pick up, my interest in this continued to grow and I just wanted to really understand the AI space and whatever. And so.
picked that back up, worked on that for about two years and then, started working on Phonely, which, you know, at the time was basically we’ve kind of, we’ve weirdly enough stayed on the same trajectory the whole time. We’ve like, you know, there’s been deviations, but we kind of felt that like voice agents back before, like voice AI was even a thing. We were like, Hey, this is gonna, this is going to be big someday. Like this is going to be a really valuable thing to businesses. and so we started working on that. We started putting that together. we got into Y Combinator, since then have grown tremendously throughout YC, post YC, building up the team out here in San Francisco and just like learning so much about how to scale a startup and also how to make, you know, AI talk like people.
Brian Bell (04:03.33) What do you wish you knew when you started film lead that you’ve kind of learned the hard way?
Will (04:08.686) Many things. I think that if I was going to give myself some advice looking back, it would be...
Will (04:20.332) Like, if this works out, think the thing that everyone underestimates and I’ve heard before, which is like how long and how hard you have to work on something to make it actually work. And I think it’s really easy to get this kind of like short termed, short term mindset when you’re making decisions in the early days, because everything feels like it’s compounding, you know, like the decisions you make like, this customer that’s paying me $7 a month is going to churn my company’s over. And I think just like having the maybe wisdom to take a little bit more of a long-term perspective and say, hey, if we do the right things, if we hire the right people, if we, you know, do what is right to our customers, we will have, we will have success. And I think it’s just maybe just like an underlying belief that if we’re solving a serious problem, We continue to work hard, we continue to iterate on it, and then from the very onset, we try to make the right decisions for the long term and not the short term. I think that that is something that would enable us to be farther ahead.
Brian Bell (05:25.25): How did you know it was the right time to do something like this? Voice AI has really come into its own in probably the last two or three years. When did you start phone Lee and when did you decide like, this is the right technological time for this?
Will (05:39.884): That’s a good question. I mean, we started working on it before it worked, right? So that’s how we know it’s like the right time to start it is because you couldn’t like, was nothing. Yeah.
Brian Bell (05:49.044): It just doesn’t quite work, but it probably will work in like six to 12 months. It’s kind of the right time to start working on it.
Will (05:54.671) Where it was like, oh, this doesn’t work now, but we could see how it worked. kind of, there was an inherent pain point that was out there. Um, and so how did I decide it was the right time? It was personally for me, it was the right time because I was ready to start something else. Um, I knew that I was meant to be a founder and I knew that there was no other choice. And then the other thing was, is like, it seemed possible. Like it seemed like, okay, well, if you connect these things and we solve these, like we, you know, you saw like you glimmers of hope that this was going to be this is gonna be possible.
Brian Bell (06:25.824) So why voice calls, right? Because, I mean, so much of tech has been centered around text and chat.
Will (06:34.146): I think part of the reason is that, right? Because so much of tech has been centered around those things. And, and in some sense, like voice is much bigger pain point to businesses, right? Because voice scales only one to one. So you have to have one person talking to another person with this chat. You can have five to six to maybe even 10 people in a conversation. So you can scale that out much faster. Right. And then voice is also one of those things where it’s a hard problem to solve because you have human interaction to deal with and.
And that it’s kind of like when you find the hard problems to solve, those are the problems that you should try to solve because less people are going to actually try to do them. And so the voice interaction for us was always like, there feels like there’s a lot of opportunity to not just make chatbots talk, but to really make true, to optimize for performance. And we think about it as we’re in the
They just figured out that you could put books on the internet stage with voice AI, right? But now the next stage is going to be like, wow, you can optimize for conversion of these AI agents for business outcomes. And that’s what we’re driving towards is it’s like there’s so much stuff that needs to be built around this so you can make an AI agent that truly is able to outperform every human agent in X, Y, and Z field. Like we’re trying to build those tools so that businesses can really get the value because they care about that. Because if you are a doctor, for example, and you miss out on a surgery that could have netted you $20,000, it’s very important that that AI agent didn’t mess that up and didn’t say the wrong thing and sounded friendly and x, and z thing. And so that’s where we say, hey, this is the problem that will need to be solved in the future. And so we try to continue to work on
Brian Bell (08:23.906) Yeah, it’s amazing. When did you know that you had something? Like, when did you first feel the pull of the market? Was it a particular segment of the market, particular use case? What was it where you’re like, OK, we’re onto something here. Let’s lean in.
Will (08:42.22) It’s kind of been a gradual pull. I don’t know that there’s been a moment where we’re like, this was it. It was kind of just throwing a lot of different things at the wall and then figuring out what customers cared about. And the way that we would figure this out is customers would complain about a feature that we just shipped not working. And we’re like, well, that’s important. Somebody, this feature that we just put into production that didn’t exist last week, people are now complaining because it doesn’t work.
Like that feels like we should probably go more into that. And so that’s really how we shaped our product a lot is just saying like, well, I think this is a good idea. We put it in, nobody would care about it. Like, okay, well, that’s not a good idea. We put something in that we didn’t think anyone would care about, or we listened to our customers, they’re like, hey, we really want this thing, we really want this thing. And we’re like, no, why would, like just you want this, nobody else is gonna want this. And then we did it and then everybody else wanted it. We’re like, oh, we should have done that more. So it’s just kind of those.
Brian Bell (09:36.962) That’s an important distinction. I’d love to double click on that. As a PM who’s built some products that were successful, but a lot of products that nobody used, products and features nobody used, how did you... It sounds like, you know, I’d like to say like product and building product is like sculpting, but you don’t know like what the underlying statue is that you’re sculpting. You kind of figure it out as you go one like little stroke of the chisel at a time.
Would you agree with that or it because it kind of sounds like that you’re kind of listening to your customers throwing ideas out there. Lots of testing. Have you have you deprecated features as well where you kind of ship something like this isn’t nobody likes this and nobody cares and take it out.
Will (10:17.206) A hundred percent. mean, think that’s, that’s like the, art of, building a successful company is customers. Everyone’s like, listen to your customers and do what they want. Customers never tell you what they want because they don’t know what they want. Right. They say they describe their problems and you have to figure out creative ways to solve them. And a lot of times like your creative ways to solve them are not how they wanted you to solve them. Right. And so I think it’s just like this, this general iteration of like, you know, figuring out the problems trying to solve them and going off of that. And then one thing that we’ll say that really helped us, the one thing that would probably 10x our product from a user perspective is, and I recommend this to any founder, anybody who’s starting out, is to do white glove onboarding for your clients. Because then you feel the pain of your product not working if you say, charge a little extra for white glove onboarding. Not in the, like, you’re not going to, sure, you’ll make more money, but really what’s going to happen is you’re to be like, this doesn’t work. This doesn’t work. This doesn’t make sense. Why the heck did we do it like this? And then you will very quickly feel the pain of your customers and be like, this is not a right way to solve the problem. And so that’s like the shortcut that we’ve learned along the way, which is like, always have an option where we build this stuff out for them because now it’s like, you know, the product stuff isn’t like our product decisions. A lot of times are coming from our customer team that’s onboarding these clients and saying like, I don’t like this. This is bad. We should do it like this. And I think that that’s the best way to like run the organization is it should flow that direction. Like product features shouldn’t flow like top down. It should be like focus should flow top down and then product features should flow customer to engineering.
Brian Bell (11:53.218) Yeah, I like that. So who’s the ICP today and how has that shifted over time?
Will (11:59.023) Yeah, so when we first started, our very first deal was $7 a month for some lady who I don’t even think she knew what we were doing, but I spent two hours on the phone with her. If I didn’t end up with something, I was going to be very mad at myself. And so at the beginning, we were selling very small, very, very small businesses, like anybody, know, 25. A hundred bucks a month, something like that, just to answer their phones with AI. We really didn’t have a focus other than like, our objective was to say like, hey, you know, save you money by answering the phones. As we’ve continued to grow as a business, it was like, I remember that first, you know, few months of, of launching Phony, it was like, we got a $7, we got a $30, we got a $100, then we got a $500 a month subscription. And then, you know, we got a 2000, like a 3000, and then 50, like, you know, it just kind of kept going, kept going up like that. And what we found is that we really started to land with customers that had a pain point in making more money. And how does a phone help with making more money? Well, it’s around the things that I talked about already. It’s like scheduling an appointment, making sure that call is answered so that they don’t miss a lead, and then focusing on the larger businesses that maybe have this problem, not just for one person, but for 10,000 people that work under that, right? So like, it’s just try to find the bigger opportunities to solve the same problem, which is like, how do you monetize more of people that are calling in, needing your service?
Brian Bell (13:49.794) So it’s interesting. So it’s a pretty broad, almost like a horizontal platform. There’s no verticalization here. And how do you think about that push-pull between building something that’s much more broadly applicable versus like vertical specific?
Will (14:03.726) I think in some sense we are becoming like more vertical, but not on a specific industry, right? There’s, different ways that you can think about verticalization of software. One is like, okay, we’re just going to own this industry. The other one is like, we’re going to own this problem. And like for us, the problem that we’re owning is how do you make your AI agents convert? Because realistically to say like, oh, how do you make your AI agents convert in, you know, roofing is very similar to how do you make your AI agents convert in, you know, insurance and HVAC.
Brian Bell (14:30.978) HVAC, LAW, yeah, yeah.
Will (14:32.47) It’s like, all, it’s the same thing, right? It’s the same like concept that we’re trying to solve rather than saying like, it’s specific to insurance. Because the only thing that’s slightly different, and the reason why we say this specifically is because every business has their own qualification criteria. So it doesn’t matter if you’re with one roofer or another roofer, it’s like, they’re gonna ask different questions based upon what they need, what they service, et cetera. So it’s really like, that’s actually not the right, for us, for our specific industry, is that’s not the right necessarily problem to solve except for like a go-to-market perspective, which is, okay, we’re just going to focus on that. And that’s more of a sales opportunity. And we’re more focused on the product and say like, this system needs to be built and people will continue to care about this in 10, 15 years.
Brian Bell (15:19.03) How do you guys scale go to market? I mean, that’s always a question I get from founders like, I’m struggling to grow. Like, what have you guys kind of learned about growth? Because you guys have done a pretty good job.
Will (15:30.304) Yeah. it’s a good question. mean, it, you know, it’s like small things that compound. think, I think that’s really like the truth of it is, we haven’t really like figured out cold calling outbound emailing. Like we haven’t figured that out. but what we have done is try to put a decent performance of ourselves on the website. So we have a, you know, we’ve worked a little bit on SEO. We’ve done a bit of work on our website to try to optimize it. And then we go to conferences. We just try to put self in places where the buyers that we’re looking for are. And then the other thing that we’ve done that has worked out is we’ll send LinkedIn connects. And we just send blank message, LinkedIn connect. If somebody connects, it’s an ICP. Or write them a personalized note. Those are the kind of things. And then post on LinkedIn. it’s realistically, it’s probably just a lot of small things that you do to optimize for it, rather than say, this is our one thing and one hack. we try to avoid the like, I think one thing that we’ve, we’ve learned the hard way is like, got into YC. We got a lot of traffic from that when we launched is there’s, think some idea that like going viral is the solution to your problems, but going viral with a product that doesn’t work is just a whole bunch of pain. Like just so much pain and so much like negativeness that you like, you have, okay, you just launched your product and it goes everywhere, it goes super viral. You get a bunch of users, your product doesn’t work. So people churn, they leave, and it’s just a very horrible, horrible experience. And it’s much better to just focus on building a good product, getting the customers, and then making our customers grow. One of our hacks is that we’ve signed customers for like two, $3,000 a month that now pay us $70,000 a month. And so we... We’ve just learned how to build a really good product with really good customers that can scale. And we just built that relationship. So right now it’s like our relationship and our outcomes are what’s scaling more so than like some crazy, know, buy reality, go to market strategy.
Brian Bell (17:36.278) Have you, you know, a common thing I hear about AI, or like a lot of people are afraid it’s going to take our jobs, you you’re literally implementing AI that replaces human beings answering phones, right? Have you measured or noticed any changes in the organizations? I mean, you’ve obviously seen increased business outcomes. That’s why they go from paying you two to three to seven to 70,000 a month. What, you know, what kind of, what have you observed inside the businesses that you’re helping?
Will (18:07.03) Yeah. We haven’t felt that super direct impact of it, I will say. I think about it from a philosophy standpoint of saying the jobs that we’re replacing are the jobs that we’re teaching people to be robots. We’re teaching people to ask the same question every single day, over and over and over again. And I’m like, is silly. And I think if you look back in the, you know, there was time where there was a job called the calculator and that person’s job was to sit in a room and they were to do math problem. Yeah, exactly. And that’s their job. They were to do that. And so it’s like, it’s the same thing as like, should we not have invented a calculator because we replaced all those jobs of people doing. Yeah, yeah, exactly. And I think that that’s the similar, you know, the similar idea that we take into it is like we want to, as a society and as a world, it’s like, I think we should focus on, you know, taking
Brian Bell (18:42.24) And a skyscraper, yeah.
Brian Bell (18:52.522) or spreadsheets for that matter.
Will (19:05.418) Having AI do the jobs where robots can do and allowing people to do more of the things that really make people human, which is connection, which is all the different parts of that.
Brian Bell (19:16.204) Do you have any case studies or stories from your clients that you’re particularly proud of?
Will (19:23.83) I feel like all, like the thing that I’m proud of, maybe it’s not as much of like our clients. Um, I think I’m more just like proud of our team, uh, in some sense where it’s like the case studies that I’m proud of is like our team and our clients like love us. Like I just got off a call right before this with somebody who’s like, dude, I, I, I eat. Foley, I breathe Foley, I go to sleep and I wake up and I dream about Foley. Um, and it’s because it’s like, we’re saving him 80% of what he’s paying and he’s just so thankful and increasing revenue at the same time. And he’s just so thankful of kind of what we’re doing. so it’s more of just like, I’m proud that our team has been able to create something like that, that people, that this guy is just like, he’s obsessed about. I think that’s really cool.
Brian Bell (19:56.372) and increasing revenue in the process.
Brian Bell (20:11.84) That’s so cool. Have you been able to dog food your product at all? You use phone lead to cell phone lead or use phone lead to, you know, answer customer service requests or, you know, answer incoming, you know, calls of people interested in talking.
Will (20:27.192) Weirdly enough, we don’t do a lot of that. We should. We definitely should. haven’t, I don’t know if we’re not, we don’t really, we dog food our product in the sense of we’re implementing on our clients. That’s how we’re dog fooding it all the time, is just using it. And we dog food it and we listen to the customer’s calls for them. And we do manual QA. We’ve got a team of 10 people that just listen to calls and are like What’s an opportunity for improvement? How could we have made our AI sound a little bit better? Answer the question a little bit differently. Why did this call fall off the path? Right. But in terms of like, are we doing this for ourselves? No, not really. Why is that? Because we focused on product. We haven’t focused on like growth and go to market simply because like we knew that we could make our customers scale and we wanted to prove to ourselves and to the future that we can do that. And now we’re at the stage of our company where it’s like our shift is our focus is shifting, right? It’s like, we know the product works. We know it’s scaling. How do we start dogfooding our own product to get more business?
Brian Bell (21:28.674) Right. It becomes as you cross into where you guys are in terms of revenue and round, it becomes like, OK, how do we scale this? Going zero to one is one thing and the one to 10 and then 10 to 100 to a whole other thing. And it requires a different kind of mindset. What surprised you most about customers that started using Phoneley?
Will (21:51.84) Surprised us most? That’s a great question.
Will (21:59.599) I think it was just like the, when you start something that is a relatively new industry to you, you get the privilege of learning a lot about a lot of different businesses and how they operate. I think the thing that impressed me was just like how many people were kind of out there and what some of these businesses did. Like I feel like I just got a free masterclass in so many different industries and so many different verticals on what they were doing to answer their phones and what they needed done and what, like, you just learn so much about all these like weird niche things. like, you know, pest control or like plumbing or like, you know, just like random things. You just get to like understand how their business works and like get a really cool view of like how the world works in a different way than I had ever, I had ever really like experienced before. Like in these industries I would never ever be a part of but I just got to learn about it. was like, oh wow, that’s crazy that this lead is worth however much money and by answering the phone, we’re able to help that.
Brian Bell (23:09.814) How is the tech stack progressed over time? What were you guys using when you got started? Were you using some voice AI platforms? Did you hack it together yourself? And kind of how has it evolved to the present?
Will (23:24.812) Yeah, for sure. So we, in the early days we did like, we started building it ourselves and then we had a period of time where we use like some other, you know, voice providers, platforms. And then we kind of realized that the problems that we were trying to solve weren’t solved by them and we just needed to build it all out ourselves. you know, at, YC we took everything in house and we just started building it out kind of from the ground up in the way that we thought it should be built. And this is like one of these things where It’s going to, it’s a pain, I guess, huge pain to build it all yourself because you have to deal with reliability. You have to deal with latency. But we do have things as like very accurate voice activity detection. Like we’ve built in and integrated our own, you know, features to that. think that it wouldn’t have been possible. So it was a huge pain. took a lot of work, but we really like, you know, we’ve been building out the tech stack ourselves, like our own Kubernetes clusters, managing scaling, just because it, it, at some point it becomes more valuable to have it reliant on your own tech, and it’s what customers are going to pay for.
Brian Bell (24:25.26) Yeah. So I’m a roofer and I find out about you guys from another, you know, small business buddy of mine. Like, what’s the onboarding look like? How fast and what’s the onboarding flow look like?
Will (24:37.23) Yeah, for sure. So it depends on how much call your call volume is, right? So if you, you know, if you’re not doing about 10,000 calls a month, we’d refer you to one of our smaller plans. Um, that would be self-serve. We have a free plan. You can get started completely for free. Um, build up the agent yourself. We’ve got a whole help center, docs, um, all that stuff. That’s the thing one people don’t know about. It’s like, have a 200 minutes. You just get for free for month. So if that’s you, you can just use one of our templates that we’ve maybe built for another roofing industry. Just plug that in.
Brian Bell (25:01.377) That’s cool.
Will (25:05.998) Get started get a free phone number everything and just get connected right away. Yeah if you are If you’re doing maybe a little bit more minutes, or maybe you want to integration to your CRM We’d have like some other self-serve plans that you could get on so you need to connect with Google Calendar You’re gonna upgrade to one of those plans, but if you’re answering more than 10,000 calls a month what it’s gonna look like is You know we we work with you to help implement when I talk about dog food your own product We help we work with you you give us your scripts you give us some call recordings you give us what you want to have done and we work with you to implement that for you. Usually time to go live is one to three weeks depending on how much integration that you need. We work on a first 30 days is like we’ll prove the value to you. It’s just like a 30 day quote unquote pilot period and if we don’t prove that we’re going to be something that you guys need in that course of time, know, it’s like walk away no problem and then we’d move on to an annual contract.
Brian Bell (26:01.474) I should probably onboard team ignite and get a free phone number for team ignite, know So like founders like will be texting me and calling me pitching me their startups or LPs are calling and That’s really cool. What was that? What was the thinking behind the freemium offering? Did did that was that right at the beginning or is that something you guys kind of brought? later in the in the journey
Will (26:23.15) It was not really from the beginning. We, we initially had like a free months and we just like made the art starter tier for free. The reason why we did it was just like, we like end of the day is like, we want to help businesses out. Right. And like, we wanted to provide a free option to it. Um, we re we realized that we weren’t going to make a lot of, like, we’re not going to make a lot of money on those small tier plans anyways. Um, and so it’s like, well, if our option is to like gate it and not let anybody use the product or to give the product away for free. At least you can feel good about helping out businesses that are.
Brian Bell (26:55.318) Plus those businesses might grow into big businesses someday. They might just be getting started and yeah, they have a small need now, but they’ll maybe they’ll have a big need, you know, years from now.
Will (27:05.068)
Yeah, yeah, absolutely. And I guess our side is just like we just, it’s like our portion of giving back and hopefully trying to add value. think the more ways in which we can add value, the more ways that we can help. People will always think about us. Maybe they might be a small business, but their buddy might be a big business, right? And so they’re using us and they refer us, and then that’s worth it.
Brian Bell (27:27.69) How have you thought about building out the team and culture as you scale and hire more people?
Will (27:34.927) Culture has been super important to us. I think we try to really pick people that are very low ego and work hard. Those are kind of our characteristics. You don’t have to be the smartest person in the world. You don’t have to have the craziest experience or whatever, but you do have to really want to work hard, really want to win, and then not think that you have all the answers. And think that picking for those characteristics has been super helpful for us. And then as we continue to go on looking forward, looking for great engineers, looking for great people in sales, marketing, growth, et cetera, as we continue to scale.
Brian Bell (28:12.736) So what are you excited about as you look forward and in the next year?
Will (28:19.116) Yeah, good question. I’m just excited, like, over the past year, the product has really come together and started to work. And so it’s cool to get to the points where we have very clear product direction, very clear focus. And then also just to see the product working with such good uptime and to see our customers being like, OK, cool, we’re going to turn all of our calls over to you instead of like half. And so just to see like, the company continued to grow and people continuing to be like, wow, this is like a truly, truly, you know, differentiated unique experience. You guys have really thought about this and really built this. I think it’s just like exciting to get more people to use that because the thing that people don’t realize it’s like every single product decision that everything on that, every button, every pixel that we have on there, like we’ve thought about really hard. and, and to just get more people that appreciate that is, it’s going to get me excited.
Brian Bell (29:13.142) What about long term? What’s the long term vision? know, 10 years out, you’re ringing the NASDAQ bell, you’re super successful. What did you build and what are you proud of?
Will (29:29.528) I think the thing that would make us really proud is to solve this. Solve the problem of helping a lot of businesses, right? And hopefully it’s going to be to help a lot of businesses to connect more consumers with businesses in a better way, right? Because as a consumer, it’s like, hate dialing ringing up to someplace and getting an IVR system or not even getting a call at all, right? Yeah, it’s horrible. And so to contribute to making the world a better place,
Will (30:03.886) For both businesses and consumers feels like a really, really awesome goal. And I guess for us, it’s just gonna be like to try to get as many businesses using this as we can. And the way that we do that is we help businesses optimize for their outcomes, which is helping consumers get what they want, right? And that’s our goal is to put those two things together. And that’s what we’re gonna be ringing the NASDAQ bell is when we figure that out.
Brian Bell (30:27.65) Let’s wrap up with some rapid fire questions. What’s a belief you hold about building AI companies that most people disagree with?
Will (30:33.56) Go for it.
Will (30:47.444) Hmm, most people disagree with.
Well, I think that this has been, I’ve kind of got two. Number one is AI is not, like AI is tapering off and how good it can be, right? And so I think that like it’s growing, but it’s not growing at the rate that it has before. So it’s a very valuable time for, yeah, and it’s capability, right? Like it’s getting better and it’s getting better and it’s progressively getting better, but it’s
Brian Bell (31:08.938) And its capabilities, But it’s no longer getting like 10x better year over year, maybe on like some math benchmarks and some arcane last humanities last exam. But like in terms of voice, I it’s not you’re not seeing the order of magnitude increases year after year as you were like maybe two or three years ago.
Will (31:27.872) And I think it’s just a really good time for people to get in and, and try to add value to businesses. I’d say that that’s probably the big one. It’s just like, you know, it’s the knowing how to use and implement AI is a hard thing. and I think that it’s not something like all big businesses want AI, but very few people know how to implement and use them. And so I think it’s an exciting time if you’re building with AI to implement solutions for a specific type of business.
Brian Bell (31:58.241) Is that both parts? You said there were two things. Yeah, OK. What skill has mattered more than raw intelligence in your journey so far?
Will (32:00.781) Yeah, that’s basically like both.
Will (32:08.91) Just grit, not giving up, and that’s it. So it’s like, don’t quit.
Brian Bell (32:14.902) Was there a time where you thought like maybe we should quit and you just persevered?
Will (32:19.544) Probably every day for six, a course of like six months to a year.
Brian Bell (32:24.13) Was that the first six months or when was that? Yeah. Six to 12 months. Yeah.
Will (32:26.83) Yeah, not as much anymore, but for the first like first year six months Yeah, because it was like, you know, it’s like right They raise more money or like, know, so you see something or the product is not working and you’re like, man I would really like to not be doing this right now Like your customers mad at you. It’s just like Yeah. Yeah, it really is. So yes most of it
Brian Bell (32:34.786) Going 0 to 1. Getting punched in the face day after day and yeah.
Brian Bell (32:51.138) Speaking of which, what’s the hardest founder decision you’ve had, you’ve made that looked wrong from the outside?
Will (33:10.07) I think from the outside, the whole, the whole thing that we like, our whole product is like what everyone says is wrong from the outside. It’s like a horizontal product is like the amount of investor calls that I had with people that are like, you should focus on a niche vertical and I want you to create an insurance AI and I want you to go up to the CRM and I’m blah, blah, blah. It’s like, that’s like a very, that’s the path. Like that’s what everybody wants is like, yeah, that’s. Yeah.
Brian Bell (33:34.818) It’s a cookie cutter. Yeah, that’s why I asked the vertical question because a lot of vertical works. But I think I think you segment it in a different way, which which is the problem. What’s the problem that we’re solving and what’s the common problem that a lot of a lot of people have a lot of businesses have. And I think it’s a it’s an interesting way to segment the the market.
Will (33:53.295) It’s, I mean, that’s like, that’s the consistent decision that we’ve constantly made because we’ve listened to our customers more than we’ve listened to like, you know, investors or people. We’re just like, well, what you going to solve your problems? And then like the things that we do to solve their problems seems like an unintelligent decision to make if you’re like going up on that approach.
Brian Bell (34:03.169) You shouldn’t listen to us like we don’t know. Yeah, don’t listen to us.
Brian Bell (34:14.55) Because the investor has like ulterior motives, right? They’re thinking about their portfolio and they’re thinking like, well, I don’t have somebody going after the voice agent insurance space. You know, that’s what they’re thinking. It’s like, we’d love for you to tackle that, you know. So, but that may not be the best thing for your team and your product. Yeah. Yeah.
Will (34:34.53) And for the customers, know, like the ones that we care about anyways, it’s like, if they don’t think it’s useful, then that’s what matters. Not what, not what everybody else thinks.
Brian Bell (34:40.96) What’s a recent development that generally changed how you think about the future?
Will (34:55.982) Recently, I had a development that changed the way we think.
Will (35:07.406) Good question. I think so. It’s not like a specific one realistically, or like, oh, this just happened and I think this is going to change the way everything works. But I think that the thing that’s really interesting is these browser agents and how good they’re getting and how fast they’re getting. I think that those two things are really exciting because it allows us to unlock access to a lot of different systems without having to build APIs.
Brian Bell (35:35.072) Right, lot of systems don’t have APIs, right? But they have GUIs, Keyboard and mouse GUIs.
Will (35:37.614) I think that’s the thing that’s a little bit underrated when it comes to voice. I don’t see many people working on that. And I think it’s going to be cool in the future. It’s going to happen. So it’s going to work.
Brian Bell (35:51.712) What do most people misunderstand about phone calls as a communication medium?
Will (36:00.171) I think that the thing that people misunderstand is how convenient it is to pick up the phone and call something and get an answer in 30 seconds. And I think a lot of people think about phone and they think about that like, that’s a dying industry. And I think realistically it’s going to be a growing industry when people start to realize that like, wait, it’s super convenient to be like, do you guys have anything available at four o’clock rather than getting on your website, figuring out where the button is.
Brian Bell (36:30.966) Logging in. man, I got to do multi-factor authentication to log in. Okay, now where was I here? Like, do they have a thing? Okay, now I got to write them and then I got to wait for a response. And then I could have just picked up. You know what’s funny is my wife and I will argue about this because she is not one. She will go through all the hoops of not making a phone call. She’ll like, is there any possible way not to make a phone call? She’ll do it. But I’m just like, I’ll just pick up the phone, right? Because to your point, you can get an answer in under 60 seconds versus.
Will (36:50.584) Like why?
Brian Bell (36:59.564) Five, 10 minutes of jumping through UI. And then...
Will (37:02.69) And but why do think she does that? Because she thinks phone calls can’t help her?
Brian Bell (37:06.754) I think there’s like this aspect of her where she doesn’t want to put people on the spot and inconvenience them. And also she tends to be very social. if she knows, I think she knows herself well, if she gets on a phone call with somebody, she’ll like become friends with them and talk to them too long. So she’s listening, she’s probably laughing, but yeah.
Will (37:27.382) Yeah. If it’s like AI solved that. Like AI will solve that because it’s, you know, he’s like, yo, I’m not going to feel bad. I’m going to get my information that I want. And so more busy, you’re going to get more phone calls and people are going to start putting the phone calls on the front of their websites again. So you can get ahold of them. Yeah.
Brian Bell (37:46.315) Right. Call this number right now. have operators are standing by as the phrase goes. Right. What’s under discussed risk invoice. I that founders should take seriously right now. You if you’re advising like a friend starting starting in a different space and voice.
Will (38:02.722)
Yeah. TCPA that is, I think that’s not consistently talk about TCPA is a telephone protection communication act or something like that. Not exactly, no, that’s what it is. but basically it’s a regulation that, deals with outbound calling. So outbound calling is, you know, has to have specific consent in order to make this happen. And there’s some gray areas where we’ll see some people that are new to the space that want to get in and they basically want to just like cold call a bunch of people using AI. we think You know, there’s illegal now and definitely going to be illegal in the future. So I think anybody who’s approaching this from a perspective of like, I don’t think you should really approach voice AI or phone calling from a like, look at AI. can now like contact a million people just like I could do with emails because that’s all going to get, there’s going to be lawsuits. It’s going to get, yeah, that’s not going to be good. so try to solve the real problem, which is like answering phones inbound. and then only use outbound if there’s very specific use cases, which there are some that are like legitimate, but just be careful.
Brian Bell (39:04.79) like Collecting Debts, that’s another YC company of ours, Collectwise. What advice would you give researchers thinking about becoming founders?
Will (39:15.79) Don’t do it.
Brian Bell (39:17.186) Just stay in academia and just get tenure and chill.
Will (39:20.302) I kind of give that advice to everybody, Like, don’t be a founder unless you have no other choice but to be a founder. Like, you know yourself, like, there’s some people that you won’t have another decision to be a founder because it’s so hard. And even if it works out, like, it’s worked out objectively very well for us, right? And every single day, it is incredibly hard. And it sounds like, that’s so nice that it’s hard.
But it’s not, it’s romanticized, but it just sucks, man. You don’t have any friends, you’re constantly working all the time, you’re constantly stressed out, all the problems, they just come up to you, and there’s really not a good motivation to do it, other than you have no choice other than to try to change the world, and that’s the reason that you’re put in this earth. And so if that doesn’t feel like you, then join a startup, that’s fun. Work on research, that’s fun, contribute in some other way, but I just wouldn’t do it unless you really feel the need.
Brian Bell (40:21.292) I have a new VC question when I meet founders now. Like, why were you put on this earth? Let’s see what they say. Because you said right at the outset here at the start was, you I knew I wanted to be an entrepreneur. I had to build something. Yeah. What do you think that is about your background where you had to do that? Right. Because you were in research, you were studying AI, PhD, all that stuff.
Will (40:46.87) It’s not like a, I think it’s just kind of like an expectation of myself as a person that came from a very young age. what a lot of people know about me is when I was like 14, I set my first world record, which was like, I had for 72 days around the world’s largest lake, you know? And so it was like, from a very young age, I had been doing these kind of like extreme things that most people could never accomplish. And so I just had this expectation and relationship with myself that like I was meant to...do extreme things that people can’t do. And then I also had a pretty big passion for trying to help the world. And so what I view startups as for me is a stepping stone to amass a lot of what the tokens that the world operates on, which is money, so that I can help redistribute the chips going forward to help make the world a better place for everybody involved. And so. For me, that’s like, it’s a very mission driven thing where it’s like, I will go through all this pain because like I want to make the world a better place. And that’s like, and I believe that I’m special. Like that’s the two reasons, right?
Brian Bell (41:57.505)Yeah. And this is the biggest thing for us as VCs is trying to find people like you to invest in, And trying to spot that in a 30 minute conversation. in terms of most of my YC meetings, 15 minute conversations. Like, what makes you unique? Why are you special? What put that chip on your shoulder? Why are you so driven? What have you done that’s like special? Right? In so many words. what do you think it is about the, you have two co-founders, right? There’s three of you.
Will (42:23.312) No, just one.
Brian Bell (42:24.928) Just one. the two of you, what do you think makes you guys, we’ve learned a lot about you, but what about the two of you? Like, what do you think, what about the two of you has made you guys successful?
Will (42:36.142) Yeah, mean, think Nisal is incredibly intelligent. He’s just very, very smart. He’s good at a lot of things that I’m bad at, which is like most of the detail oriented stuff. Like I’m the very big picture guy, right? Where I say like, okay, this is what we’re going to do. This is roughly how I’m going to do it. And he’s good at making things happen.
Brian Bell (42:55.84) Yeah, gotcha. Well, last question. What’s a question about AI or startups you think more people should be asking right now?
Will (43:14.454) Hmm. Good. I think the biggest question that people should be asking is where is the value? If you’re creating a startup and you’re, you know, a rapper on top of open AI, that’s fine. But like really where’s your value and how is that value going to be communicated going forward? And I think if you really understand where your value is being created, then you can do a great job at pitching that to businesses.
Right now we’re in this period where people will just kind of like throw money at AI because they don’t really know what’s happening. But at some point in the future, people are going to know what’s happening. They’re going to understand. And you have to really kind of always think of like, how am I getting closer to value from end businesses and consumers, whatever you’re going for, and how do I double down on that feature?
Brian Bell (44:05.292) Well, Will, amazing progress. Great to get to know you better. I learned a lot. I’m sure entrepreneurs out there will learn a ton from this episode. Thanks for spending some time with us.
Will (44:15.599) Yeah, of course. Thanks so much.
