How Metropolis Condominium Management Cut Call Handling Time by 60% and Gave Their Team Their Nights Back — With an AI Voice Agent

About Metropolis Condominium Management

Metropolis Condo Management is a multi-million dollar HOA and property management company operating across Maryland, Virginia, and Washington, D.C. They manage 185 communities — condominiums, townhomes, and homeowners associations — serving between 45,000 and 50,000 homeowners. Their business runs on responsiveness. When a homeowner has a question, a complaint, or an urgent problem, it’s Metropolis’s phone team that picks up.

At the time we started working together, that team was fielding between 100 and 150 calls every single day.

The Problem: A Phone Team Buried in the Same Questions, Day After Day

Martin Lobb, the founder of Metropolis, is the kind of operator who had already done a lot of things right. His team had built their own internal software, Community Watch, to track every call, email, and document through a unique tracking number. They’d moved away from expensive U.S.-based phone staff and built a dedicated team in the Philippines — 43 people who genuinely cared about the homeowners they served and had mastered Metropolis’s internal processes. By most standards, Metropolis was ahead of the curve.

And yet, the phones were still winning.

“We have way too many phone calls.”

Martin Lobb, Metropolis Condo Management

With 100 to 150 inbound calls per day, Metropolis needed seven people staffed on phones at all times just to keep up. And the problem wasn’t the volume alone — it was the nature of the calls. The vast majority were repeat inquiries: homeowners asking about their balance, where to send a payment, how to submit a color change request for their home, or how to write a letter to their board of directors. The same questions, with the same answers, handled by trained human beings who had to look up the same information, remain calm, and repeat the same explanations dozens of times per shift.

“A lot of the calls are rote — people are looking for the same questions over and over.”

Martin Lobb

The pressure didn’t stop at 5 PM, either. Metropolis had an after-hours answering service to handle emergencies — a necessary expense, but a chaotic one. True emergencies at Metropolis fall into exactly three categories: fire, flood, and lack of heat. Simple enough. But the answering service couldn’t always tell the difference between a genuine emergency and a homeowner calling at 2 in the morning about a parking pass.

“We would get calls at 2 o’clock in the morning because someone would be calling about a parking pass. And the answering service was being held hostage, so they would call the person on call.”

Martin Lobb

Team members were being woken up in the middle of the night for non-emergencies. Morale was taking hits. And no matter how much Martin invested in staffing, training, and process, the ceiling on what his team could do was set by how many calls they were forced to answer manually.

Martin knew there had to be a smarter way. He just needed to find it.

The Solution: Meet Pixel, Metropolis’s AI Community Receptionist

Martin initially came looking for an AI marketing assistant — something more polished and professional to represent Metropolis to the outside world. But the more he and I talked, the clearer it became that the real leverage wasn’t in marketing. It was in the phones.

What Metropolis had was a call operation where roughly half of every interaction followed a predictable, structured script. The questions were known. The answers were documented. The processes were already in place. That’s exactly the kind of environment where an AI voice agent doesn’t just fit — it thrives.

We built Pixel: Metropolis’s AI Community Receptionist. Pixel speaks with homeowners in natural, warm, conversational language — and, as Martin quickly discovered, she has a sense of humor.

“Some of our introduction recordings are, ‘Would you like to hear a joke?’ And when I did that, everyone was like, Oh God, don’t do that. People are going to hate that. They love it. It makes it more personable. With Pixel, they want to talk with her.”

Martin Lobb

Pixel handles the full range of routine homeowner inquiries: balance questions, payment instructions, maintenance requests, HOA rule clarifications, document submissions, color change approvals — the entire catalog of repetitive calls that had been consuming the team’s time. She answers in English and Spanish, handles up to 15 simultaneous calls without a single caller reaching voicemail, logs every interaction automatically into Community Watch, and sends homeowners an email confirmation once their request is filed.

And critically: if a homeowner wants to speak to a human being at any point, they can. Pixel never traps anyone in a loop. The bridge to a live team member is always available.

For after-hours calls, Pixel applies the same logic Martin’s team already used — she just enforces it consistently, every time, without fail. Fire, flood, or lack of heat: escalate immediately. Parking pass at 2 AM: handled gracefully, with clear alternatives offered and no human woken up in the process.

“With Pixel, it’s very different. Pixel knows exactly what the specifications are. She doesn’t get involved. She doesn’t argue with the client. She’s very friendly, but very assertive as well, saying, ‘Unfortunately, I’m not able to assist with this. Here are your alternatives.’”

Martin Lobb

What It Took to Build It: Training Pixel Together

One of the most common concerns we hear before a project like this begins is how much work it will take from the client’s team. The honest answer is: some — and it’s worth every minute of it.

Martin approached the training process the right way. He assembled a committee of his phone team members, had them map out every question type they encountered, and used Metropolis’s existing call tracking data — which already logged the name, address, phone number, and reason for every call — to build the knowledge base. The groundwork was already there. We gave it structure.

“Remember Smita? She deliberately tried to break the system.”

Martin Lobb

One team member’s job was specifically to find the cracks — to ask edge-case questions, phrase things in unexpected ways, and push Pixel until she stumbled. That adversarial testing approach is exactly what separates a brittle AI agent from a robust one. Every failure she found in testing was a call answered correctly in production.

Martin’s assessment of the process: surprisingly manageable.

“For something as complicated as we had, your ideas and your input just made it a really fun experience.”

Martin Lobb

“I Went From Wanting a Robot Voice to Hiring Another Person”

Before going live, Martin had assumed AI would sound robotic — and honestly, he’d initially wanted that, thinking a retro, futuristic robot voice would be a fun marketing gimmick. What changed his mind was experiencing what Pixel actually sounded like: human, warm, and disarmingly easy to talk to.

“People were really receptive. So for me, I went from thinking of a more robotized voice and marketing to actually hiring another person. Because as far as I’m concerned, I can call Pixel and talk to her about anything.”

Martin Lobb

This is one of the most persistent misconceptions about AI voice systems: that customers will feel tricked, or that they’ll reject the interaction because they know it isn’t human. Martin’s experience is exactly the opposite. Metropolis is transparent that Pixel is an AI — and homeowners have responded by wanting to talk to her. They call hoping they’ll get Pixel. They test her. And she’s always ready.

“There are times when I’m talking to Pixel on testing her that I almost forget that she’s not a real person.”

Martin Lobb

The Results: 60% of Routine Calls Gone. Nights Restored. A Team That Actually Gets to Solve Problems.

The numbers from Metropolis’s early implementation tell a clear story.

60% time savings on basic phone calls. Pixel now handles the majority of routine homeowner inquiries — the kind that used to occupy seven team members all day. That’s not a marginal improvement. It’s a structural shift in how the phone operation works.

30% of after-hours calls completely eliminated. The calls that shouldn’t have been escalated in the first place — the parking passes, the general inquiries, the things that had no business waking anyone up — are now resolved by Pixel at the first point of contact, without involving a human at all.

Zero overflow. Before Pixel, missed calls meant voicemails. Voicemails meant complaints. Homeowners asking why nobody answered. Now, Pixel answers every call the moment it arrives, regardless of how many are coming in at once. Up to 15 calls simultaneously — without a single one going to voicemail.

An end to employee burnout. This one matters more than the numbers suggest. Being asked the same question for the fifteenth time in a day wears people down. It erodes professionalism. It costs real money in turnover and morale. Pixel doesn’t burn out.

“You don’t have employee burnout. Pixel is always friendly. She’s always funny. She always has a joke. So if you’re asked 15 times what my balance is, and you have to go through all of the steps — you get burnout. But with Pixel, she’s very friendly. She’s very comfortable. She knows what she’s doing.”

Martin Lobb

A team freed to do higher-value work. The phone team members who spent their days fielding basic inquiries are now available for the complex calls: the ones that require judgment, empathy, and real problem-solving. Community counselors have lighter workloads. The whole organization runs at a higher level.

– Martin puts it this way:

“My crew of seven people can now move to higher level problem solving. And they can have an easy, enjoyable communication with the homeowner. So the homeowner feels important and valued and listened to while Pixel is able to handle the undercurrent.”

Martin Lobb

And where is Metropolis headed from here? Martin has a clear vision. He wants Pixel to eventually initiate outbound calls — following up with contractors, confirming emergency resolutions, reaching out to homeowners proactively when needed. The foundation is built. The protocols are being laid, step by step.

“What you’ve taught me is that if you do step by step by step, they’re just as good as us — and they don’t make as many mistakes. So I have real plans for Pixel.”

Martin Lobb

One More Thing: This Works for Smaller Teams Too

If you’re reading this and thinking “that’s impressive, but we’re nowhere near Metropolis’s size” — Martin has something to say to you directly.

“Honestly, you’re too small to realize an opportunity that’s coming your way. And if you’re going to be a business, you need to have a vision.”

Martin Lobb

Metropolis is a multi-million dollar operation. But the reason AI worked for them isn’t their size — it’s that they had repetitive calls and documented processes. Every HOA and property management company receiving more than 80 calls a day has those. The discipline of building Pixel forced Martin to formalize the knowledge that had been living in people’s heads and turn it into something that could work for him around the clock.

“Whether you’re a two-man operation or you’re a large company like myself, if you take the time to invest in creating that person… you’re going to discover that they make you better. Because they force you to look at your formulas and your protocols.”

Martin Lobb