How Zest Properties Uses AI to Handle 5,000+ Monthly Interactions Across 1,000+ Units

Before, we managed our leads with a lot of hopes and dreams. We provided our SOPs and information and uploaded it into her brain. We learned pretty quickly. We actually had a flood situation, and the AI was able to talk the resident through shutting off the water while emergency services were dispatched. We just started playing around with Ava, the AI integration for making changes, and it has been game changing. We have an in-office leasing assistant who was responsible for filtering all of those calls. Our front desk associate would assist with those calls when there was overflow, but it was manual labor before. We were receiving maybe one or two people coming to the open houses before with our social media marketing and our website marketing. Now that we have the AI interacting, we are seeing maybe five or six people showing up in addition to our regular traffic.

My name is Vanessa. I am with Zest Properties. We are a residential and multifamily property management company. We are based out of Metro Detroit, and we have been around since 2018. We have a very high call volume, and we have a lot of repeat calls where we give the caller the same information. Those were things we wanted to automate while keeping the experience interactive for the residents who were calling, so they still felt heard even though it was the same information over and over.

We are newer to the delinquency side. We have had our first full run-through now, and we were able to do all of our delinquency calls with the click of a button. What would normally take us all day to call through our delinquency list was done in five minutes, which saved so much labor. We would spend eight hours a week doing outbound delinquency calls, and now that is done with the click of a button. We are saving substantial time every month.

We have the API set up so it can read our listings and provide information on our open houses. A lot of the phone calls we received were questions about when the next open house was and what features the house had. We can filter out a good percentage of those incoming phone calls with AI and still complete an intake form, like a guest card, so we have all of their information in our database and can follow up with them.

Before, we managed our leads with a lot of hopes and dreams. We would take phone calls during the day and return voicemails for calls that came after hours. We did not really have any way to catch those. We have an in-office leasing assistant who was responsible for filtering all of those calls. Our front desk associate would assist when there was overflow, but it was manual labor. Now we can bypass a lot of that manual labor, review our intake forms, and go from there. We were losing a lot of leads based on volume. A lot of people would call after hours and not leave a voicemail. Now we have more complete data.

We were receiving maybe one or two people coming to the open houses before with our social media marketing and website marketing. Now that we have the AI interacting, we are seeing maybe five or six people showing up in addition to our regular traffic. We are also able to direct and filter our maintenance calls much better. Non-emergency calls used to be transferred, but the AI system can filter those out. There have been a few true emergencies where residents called and the system directed them to the fire station, if there was a fire, or another safety representative. Before, they were just trying to call their management company. Hearing those calls and watching those interactions has protected the asset by getting faster responses from emergency services and has helped our residents with communication.

We used to have an emergency line. We rotated which technician was on call to answer it throughout the night. We captured most of those calls, but every now and then we missed one. Now we have a 100 percent capture rate on emergency calls, and the AI directs them accordingly. We actually had a flood situation where the AI talked the resident through shutting off the water while emergency services were dispatched.

For example, a resident called about tuckpointing, and the AI was able to explain how the weather affects our scheduling for that work. It was really interesting to hear the AI provide the maintenance information. It prevented questions from reaching our maintenance department, so it filtered that labor again.

Adoption of AI has been a change-management initiative that we have been rolling out. Our staff had some hesitation about handing over control. We were worried about customer service levels declining, but we have actually seen an increase since making the transition. We also had a lot of concern about regulations involving the Fair Debt Collection Practices Act and fair housing. Once we rolled it out and the team started hearing the calls and seeing the interactions, adoption across the organization greatly improved. Everybody on the team is excited about it. They bring suggestions and things they want to change in the AI.

The onboarding process was not as hard as we thought it would be. We provided our SOPs and information and uploaded them into her brain. We set up our intake forms, how we wanted calls directed, and the general information. It was a pretty streamlined rollout. We had a wonderful representative. She was fantastic through the whole process and has been available for any questions. If I see something going wrong, she is just the click of a button away. Having that communication line has built a lot of confidence.

We have just started playing around with Ava, the AI integration for making changes, and it has been game changing. I can type in plain English what I want to happen. She goes in, makes the changes in the system, reroutes everything, and I click accept. That has been huge. I uploaded the chatbot into our WordPress site. It was really easy to copy the code over, and it started interacting right away.

We are getting leasing leads, and people ask maintenance questions and troubleshoot through the chatbot. People inquire about open-house dates, and it can route them to the open house. We do see that traffic come through.