Ryan Young leads one of Ohio’s top-ranked real estate teams, the Young Team, which his parents, Terry and Jeffrey, founded in 2003. Ryan also admits that he often can’t get his agents to make phone calls.
TAKE THE INMAN INTEL INDEX SURVEY
Ryan Young
That’s not a Young Team problem, though. It’s a real estate problem. Teams spend thousands on marketing platforms that generate high-intent leads, then watch those leads transact with competitors because no one followed up.
Young knows this as both the CEO and co-founder of Fello, a proptech platform used by about 3,000 real estate teams nationally, and as a customer of his own product.
“Fello is amazing. It creates a significant amount of engagement in my database,” Young said, paraphrasing the feedback he’d heard for years. “Then you’re telling me that these people are actually listing their homes with our competition?”
Fello’s answer to this problem is Felix, an AI agent the company is positioning not as yet another voice bot, but as what Young calls an “agentic operating system” for real estate teams.
Felix calls, texts and emails leads autonomously, adapts its outreach strategy in real time as new data comes in, and hands off qualified prospects to human agents, sometimes while still keeping the prospect on the line.
The product is currently in beta with about 20 teams, and Fello says it plans to launch publicly within the next 30 days.
‘Do you want us to call them for you?’
Fello launched in 2021 as a direct-to-consumer iBuying concept, initially called FlashHouse. Around the same time, interest rates began to climb, and Young said they realized iBuying wasn’t scalable for them.
“In October 2022, we pivoted quickly to a SaaS, subscription-based model, selling our platform to agents all over the country,” Young said. “Agents would sync their database into our platform, and we’d market seller messaging on their behalf: get a home value, get a cash offer.”
The model grew quickly. Young said they currently have about 3,000 real estate teams on their platform and about 40,000 users.
“And what was interesting over the past three years, we’ve done a masterful job at creating all of this engagement through email marketing in these people’s databases that they really were just very dormant and very quiet and kind of sleepy,” Young said. “And we woke them up, right? Much of database engagement is predominantly on the seller side.”
Fello had a problem, though. Many of their customers loved the platform and said it was creating engagement. But then the prospects who were engaging were listing their homes with their competition. Fello’s customers couldn’t get their agents to make the phone calls and consistently reach out to these prospects.
“We heard this feedback so consistently over the past couple of years that we joked, ‘Do you just want us to call them for you?'” Young said. “And honestly, we finally realized, maybe there is a way we can do exactly that.”
Felix is the answer to that joke taken seriously. The AI agent ingests Fello’s underlying contact data — property ownership history, equity positions, mortgage type, AVM estimates, MLS activity — and builds an automated outreach strategy for each contact. When a contact’s status changes, the strategy updates automatically.
The demo Young gave Inman illustrated the specificity.
In one case, Felix detected that a contact had recently purchased a new property while still owning a home he’d held since 2003. Felix’s outbound text referenced both properties by address, and the prospect engaged. Within hours, Felix had called him and transferred the conversation to Young’s team as a live listing appointment.
Adjusting strategy in real-time
What separates Felix from the wave of AI calling tools entering the market, according to Fello co-founder and Chief Product Officer Tom Schrader, isn’t the voice but the connective tissue beneath it.
“It’s not the fact of even just making the call, because nobody’s going to make the calls anyway,” Schrader said. “It’s to get a result that requires zero effort.”
The setup process involves Felix scraping the real estate agent’s website to understand their markets, then beginning outreach without the agent configuring scripts, cadences or triggers. Schrader said Felix will stay with a contact for months or years, adjusting its strategy as intent signals change.
One example Young showed: a prospect who mentioned using a 401(k) for a down payment during a call with Felix. Felix subsequently generated and sent a personalized email outlining 401(k) withdrawal strategies for homebuyers, not from a template, Young said, but composed in the context of that specific conversation.
“When people see this, they’re like, how did you guys have a template built for 401(k)s?” Young said. “I’m like, there are no templates.” In other words, Felix had customized the email in real-time exactly for that scenario.
The all-important handoff to a human agent
The handoff from Felix to a human real estate agent is the part of the product Fello is most focused on getting right.
Most AI calling tools, when they qualify a lead, put the prospect on hold and try to transfer to a human agent. It’s a process, Young says, that produces roughly a 70 percent drop-off rate before the agent picks up.
Felix instead keeps the prospect in a live conversation while simultaneously briefing the agent on a separate line, then bridges both parties when the human real estate agent confirms they’re ready.
Young’s team received 30 handoffs on a single Monday during the beta, he said. The platform’s notification dashboard showed handoffs arriving minutes apart in real time during the demo.
“You’re focusing on the bottom of the funnel, not the top of the funnel,” Schrader said. “And that’s all you have to focus on.”
On compliance, Fello uses its existing email platform to capture TCPA-compliant consent before Felix initiates calls or texts. The company built its own consent-capture tooling rather than using a third-party service, Schrader said.
Felix is also scrubbed against the Do Not Call registry, and the platform has state-specific settings for two-way consent requirements and pre-recorded call disclosures.
‘Have you ever seen the movie Her?’
Getting the voice to sound human required more than swapping in a premium text-to-speech engine. Schrader said Fello layers proprietary prompting and what he called “disfluency engineering” — deliberate pauses, hesitations and inflection cues — onto third-party voice engines, including ElevenLabs.
“We focus on how we get better inflection and better emotion, better pausing. The natural disfluencies that we need to introduce into the cadence of speech to make it feel human,” Schrader said. “That is not something that just happens without being intentional about it.”
Schrader joked that, as founders, he and Young were naturally very “obsessed” with getting it exactly right. “We went super granular on all the little things that make a human sound human,” Schrader said.
In the demo calls Young played, two separate prospects told Felix mid-conversation that they thought they were talking to a real person. Both continued talking after being told they were, in fact, talking to an AI agent.
On one of the calls, the prospect stopped after a minute or so and said, “It seems — it certainly seems like I’m talking to a real person, but just something about this makes me think it might be an AI call. Am I talking to a real person right now?”
Felix replied, “Oh, you caught me. Yeah, I’m actually a digital assistant with the Young Team, so you’re talking to an AI, but a really friendly one. I’m here to help get the conversation started, gather some info, and then connect you with one of our human experts who can dive into the details with you. Does that work, or would you prefer I have a real person call you back directly?”
The prospect, who sounded less annoyed and more impressed, said, “This is one of the most sophisticated AIs I’ve ever talked to. I thought for a while I was talking to a real person.”
In another call played for Inman during the demo, the prospect said, “Can I just tell you something? If you had told me you were an actual person, I would have believed you.”
The Felix AI agent confidently replied, “Honestly, I’ll take that as a huge compliment. I know I’m digital, but I really do try to keep things feeling natural and actually helpful. [That] made my day.”
The prospect replied, “Even though you’re not real.”
Felix, quick with a response, shot back, “Hey, I’m real enough to have this conversation with you. But seriously — do you want me to have someone reach out about that appraisal question, or are you feeling pretty good about where things stand?”
LISTEN TO THE CALL BELOW
Not everyone, of course, wants to talk to a bot. But Young’s answer to that objection is the call length.
“These aren’t one or two-minute calls. Some prospects are talking to AI for 15 to 20 minutes. And the reason is context,” Young said. “We built the platform on rich data: home equity, mortgage types, properties owned. So when the AI reaches out, it’s not a cold call. It has enough information to have a genuinely meaningful conversation.”
AI voice adoption is accelerating fast. Conversations with AI can still unsettle some people due to the “uncanny valley” effect of hearing something that sounds almost human but not quite. But familiarity is winning out. Millions of consumers have grown up talking to Siri and Alexa, and are now increasingly turning to the voice modes built into ChatGPT, Claude and other AI platforms.
“Have you ever seen the movie Her? This is where I think our world is heading,” Young said. “I think people actually are starting to prefer talking to AI.”
‘I can’t imagine going back to the old way’
Young said Felix has set approximately 60 listing appointments for his team during the 90-day beta period. The Young Team has been using the Felix product and acting as a live laboratory to test and improve it.
The comparison Young keeps returning to isn’t other AI calling products. It’s the inside sales associate model that many top teams have built their operations around.
“I can’t train an ISA or an agent in 90 days,” Young said. “I’ve had ISAs with me for a year, and they still don’t know the homes that we’ve sold in the past. Felix knows all that stuff, and he knows it instantaneously. I don’t have to fire Felix and then go rehire a new ISA.”
But then Young continued, “Here’s the scary thing. It’s great for Fello. It’s scary for our industry. I can’t imagine going back to the old way.”
Young said that if he lost Felix, he would suddenly have to hold his agents and ISAs accountable for all the calls and texts Felix had made.
“The biggest challenge that our industry has is that we’re not really strong operators. We’re salespeople. And we grow big companies,” Young said. “And we struggle to create a consistent process. And there’s a lack of accountability.”
For most real estate professionals, Young said, they “want to be out and about selling, and we struggle for discipline.”
“And so we’ve had this vision to build this agentic operating system of like, how do you lean into what’s happening in the world, this agentic future that’s happening, and help that be your operations and consistency so that you can ultimately be out selling homes, doing the things that you love to do,” Young said.
