Speaker 1 (00:00):
Most real estate AI conversations start and end with the same question. What can the agent stop doing? Tiffany Gelzenis thinks this is the wrong question. As a real estate team leader, follow-up boss realtor and residents and three-time guest here on Real Estate Team OS, Tiffany makes the case that AI's highest use isn't in what it takes off the agent's plate, but in what it does for the client. Here in this episode, you'll learn how the difference between generative AI and Agentic AI changes the way that you deploy every tool in your stack. What's hiding in your database's nurture black hole and how behavioral data surfaces it, why your AI is only as good as your CRM data and what you can do about it right now. And the consumer experience trend that separates AI that matters from everything else. Enjoy your time with Tiffany Gelzenis right now on Real Estate Team OS.

Speaker 2 (00:53):
No matter where your business is today or where you want to take it, you'll get there faster and more profitably with an operating system. Welcome to Team OS, your guide to starting, growing, and optimizing real estate team. Here's your host, Ethan Butte.

Speaker 1 (01:07):
Tiffany, I love working with you. I'm excited to have you as a three-time guest and one of our first guests. Welcome back to Real Estate Team OS.

Speaker 3 (01:15):
Thank you, Ethan. I love working with you too.

Speaker 1 (01:18):
And we're going to get into some of the things we've worked on together. We're not going to do a lot of background. We will do a little for folks that want to go deeper into Tiffany's background because you love this conversation. Episodes five and episode 76. Those are both going to be linked up right down below. But I'm going to ask you, Tiffany, something I haven't asked you in a couple of years now. And that is, what is a must have characteristic of a high performing team?

Speaker 3 (01:39):
While my opinion probably hasn't changed of what the must-have characteristic is, this is probably going to be different than what I said way back then. I'm going to go with consistency today because you can have all the tools in the world. You know I love my tech and I love innovation, but if you're not consistent with how you use it, then you might as well not pay for it. So consistency is my answer.

Speaker 1 (02:04):
What gets in people's way?

Speaker 3 (02:06):
Oh, usually themselves. Yeah, they get in their own way. Sometimes it's a mindset issue where people are just not thinking through things the right way. A lot of it is time management. For the most part, it's time management. They allow things to distract them and they don't focus on what is effective every single

Speaker 1 (02:26):
Day. Team leaders were nodding when you said consistency, and I'm sure they thought of a third or a quarter or worst case scenario, half of their agents and they're like, "Yep, that's a thing." And in the team environment, of course, the team leader or the ops leader is generally introducing new tools, new tech, new processes, new systems. Any tips on adoption, whether it's mindset or whether it's behavior, like any tips for someone who's like, "Yeah, I struggle with that. Some of my agents aren't as consistent as I need and want them to be. " One or two things that have been helpful for you in getting agents to do the thing.

Speaker 3 (03:00):
One is simplify it. So your agents do not need to understand all of the ins and outs of how every single piece of technology that you have works. So simplify it for them. For me, a lot of times what I do is if I'm deploying a new technology, I won't necessarily even tell the agents that I'm deploying a new technology. I'll just add the different priorities to the existing workflow that we already have. When you change things up too often, there is this whole analysis paralysis piece that comes into play. So don't over complicate it for your agents. Simplify it for them. They just need to get in there and make the calls. Your job is to prioritize those leads, make sure you have good leads coming in, but to prioritize those opportunities so they're calling the right people at the right time. That is the best way for them to stay consistent is when they're feeling like this is an effective use of my time.

Speaker 3 (03:59):
When they know they're calling the right people, they're more likely to make those calls and to do those activities.

Speaker 1 (04:06):
Man, I love that. So I feel like you're kind of foreshadowing a little bit. We're definitely going to be hitting that a little bit more. The reward for the agent of doing things that are effective, they don't necessarily need to know that anything changed. You're just changing the process, the workflow. The prioritization was a key word that you used there. Before we get into that, I would love for you to characterize the Gelzenis group however you would like. What was going on when you started it? How long you've been doing it? What does it look like? Who do you serve? Anything you want to share.

Speaker 3 (04:33):
Sure, sure. So we have been operating as a team since 2017. We've had some iterations. You can hear about those back in episode five. I mean, I think that's definitely how any team works is they go through some changes and they fine tune and they figure out what works for them and what works for the owner or the team lead. What's changed in the last two years is I still have about the same size team. So we're hovering at that eight to 10 agent mark. In the last year and a half, really, I've increased the work that I do with Followup Boss and now Zillow. And so I have taken a leader on my team and put him in kind of the seat of day-to-day leader agent accountability, sales and operations. And I'm focusing more on the tech optimization, which works really nicely with what I'm doing with follow-up boss as well.

Speaker 3 (05:33):
We hover around that 100 to 120 transactions a year. We did take a little bit of a dip last year. I call it a transition year when we moved Joe into that team leader role. It's a constant thing when you take somebody that's a top producer out of production, whether it's you as the team leader or it's somebody that you're elevating into management and have to kind of replace their activity. So we did see that happen last year, but we are on the upswing now, just kind of getting in there and doing the thing. But I like to characterize our agents as professional consultants more so than salespeople. I like to have productive agents. And so my goal is always focusing on the per agent productivity versus the quantity of agents on my team. So our goal is increasing that every single year. And I tend to have agents that stick around for that reason because we're just pouring into their business to help them become more productive.

Speaker 1 (06:32):
Awesome. One follow-up question here, but I'm going to qualify for folks watching and listening. Tiffany is realtor and residents at FollowUp Us. She's been doing that for years. That is how I met her. And so that is what takes up, I mean, generally speaking, half of your professional life, maybe more. Probably more.

Speaker 1 (06:50):
We don't need to get into that part of it, but what was it about Joe and you? How did he communicate to you or how did you approach him to say him either to acknowledge in some way, "I'm ready for more." What might that look like? Or you saying, "I think you're ready for more. Here's what I think that might look like. " Because obviously it's not to be realtor and residents at a software company, but most team leaders are looking for opportunities to put themselves in their ideal role and that might involve a move like one that you made. I'd love any story or kind of lessons from that transition besides you need to replace that person's production because they're probably a productive agent.

Speaker 3 (07:33):
Yeah. Yeah. So it was really natural for Joe to move into this position. One reason is because he had a great story where he joined our team. He was not necessarily producing at the level that he wanted to produce. He adopted our process. He, I call it followed the recipe. He followed the recipe and he saw his production increase year over year while he was on my team. So Joe's been with me now for six years, just about six years. And it took him, like a lot of agents, we're entrepreneurs, so we think we know how we want to do things. And it took him a little bit of time to be like, "Okay, let me just listen to Tiffany and what she's telling me and just plug into the system." And when he did plug in, it was just amazing. It was amazing. And he embodies our culture and he was my natural go- to for situations where if I was going to be out of town and I needed somebody to run a team meeting or if I needed to get a new agent up to speed, he just wasn't my natural go- to for that.

Speaker 3 (08:44):
And he did express interest in wanting to be more involved in agent training and in kind of a management type role. So it was pretty easy just to plug him in there. We have very complimentary traits, and so that works really well with having somebody in a management position. And honestly, I have two really great leaders on my team. I can't not call out Tara as well. She's been on my team for a long time and is also just embodies our culture. She's my go- to when I have questions, but Joe was the natural person that wanted a full-time position to move into more of a management role and really has the patience and the heart for training agents and the story to tell them of, "Look, I did the thing and it worked. So just plug in and do and stay consistent and it'll work for you too." So that's the reason why.

Speaker 1 (09:42):
I love that in part because when I hear you mention consistency, I'm sure there's an agent in your past who you're like, "I see the potential for this person, but they're just maybe not doing the things." And a lot of folks will say, "Run the play that's called." I've heard that one a number of times on a similar context is like, "We built this thing. The reason you would join this thing is that we built this thing. It's all laid out. You just plug right in. We've got the tools, we've got the systems, we've got the ... And if you're a good fit for this thing that we built, do the thing, run the thing, use the recipe. That is why you joined a team." And so I love that he brought that to life for you and that he's able to kind of continue moving it forward.

Speaker 1 (10:27):
To transition a little bit to some of the work that you do as realtor and residents at Follow Pause, it's a number of different things, but you produced all the conversion summit content last year. This year we ended up doing Fub Live, so it's a bit more Fub product oriented. I'd love for you to speak at a high level. When you learned that we were going to do these all day follow-up boss focused events, how you approached, I have a whole day to teach people in a room full of people that range from I'm brand new or I'm considering follow-up boss all the way to, I'm a pro like Tiffany. I've been using it for years like Tiffany. I know all the things and I push it in different. It's like, that's a really difficult room to serve. How did you think about Fub Live from day zero?

Speaker 3 (11:13):
Yeah. So to be perfectly honest, on day zero, I felt a little overwhelmed for exactly the reasons that you just spelled out. When you have a room of people that have varying degrees of skills, some people don't even know what a smart list is, whereas others are running a highly automated business and trying to provide value to the entire room was definitely an overwhelming task to start with. But we're asked all the time for more training. People want us to get in there and show them how to do things. So when I approach the content for Fub Live, I kind of started at the beginning. I mean, the assumption is you've already got leads in your database, so I'm not teaching you how to import them, but what I am doing is I'm starting at just the basics. Like for example, smart list, that's the basics.

Speaker 3 (12:09):
How do you run smart lists? What are the key components to a smart list to make it successful? And then going from that to lead flow and lead management to automations to sending out, creating efficiencies with templates, but also incorporating into each of these sessions the basic skillsets with some key tips and tricks that maybe an advanced user might not know about. And I'm going to tell you what, I am the realtor in residence for follow-up boss and I am supposedly an expert and I learn things every day that I didn't know existed or I didn't know that that was a different way to do things. So there are always tips and tricks to be learned in follow-up boss. So that was kind of the goal was to start at the beginning and then build on it. And we end the day actually with a session that goes through someone's entire workflow and how they've incorporated their workflow, this marketing plan that they have and used these various aspects from smart lists to automations, to text templates, to the smart messaging, and how they're using it and deploying it in an entire workflow.

Speaker 3 (13:31):
So everyone can kind of see how the pieces and parts work together versus a training session that is more along the lines of click this button and then insert this thing. It's really showing how to work it all together into a plan.

Speaker 1 (13:45):
I love it. So you lay out the pieces, you talk about them individually and then you bring them all together. The closing session ...

Speaker 3 (13:53):
Oh yeah, the actual closing session. Yes.

Speaker 1 (13:55):
Yeah. That's where I want to spend like, I don't know, at least 20 minutes with you because you put that together. I forget exactly what you titled it, but it's AI. I want to go high level. I want to speak to people that, whether they use follow-up boss or not, this will be interesting and helpful, at least for like mental framing around where this is going. If you do have features or cool things to mention, feel free to do so. But that closing session on AI, what was your goal with that session? How did you frame that up in your mind to send people off at the end of a day that's focused on AI, tech and automation?

Speaker 3 (14:31):
Yeah. So I mean, AI is definitely a buzzword, right? We're all talking about it. There are some agents that are afraid of it. There are other agents that have embraced it and it's just moving so quickly. So the session was focused on the future of real estate, but the future of real estate when it comes to technology that matters or innovation that matters, because we can get so distracted by shiny objects and we want to go and build something in claude code, well, are you building the right thing or focusing on the pieces and parts of AI that we understand without understanding the backbone behind all of it. So it was really an exercise to incorporate all of the various iterations of AI and what we're comfortable with right now, and then how we're moving forward, and then also addressing how Followup Boss is innovating alongside with AI because things are moving really quickly.

Speaker 1 (15:31):
I like the setup there. It makes me think of something that's been echoing with me a lot over the past six to 12 months, which is if we want real estate to be a human to human relationship based business, we need to treat it that way. And what we want doesn't matter as much as what is the best way to do this business is another way to think about it. And I think if it is to be a human to human relationship based business, we need to treat it that way. And that then puts the tech kind of in its place and allows us to have some kind of a filter for what are we doing when, why, and how, kind of as you already described. When you think about the role of an agent versus the role of AI tech automation, generally, we'll just bucket all that together.

Speaker 1 (16:19):
What do you think when we're deploying tools the right way in support of an agent, like what is the agent doing? What is the tech doing? Obviously not all the way into all the details hour by hour of a week of a real estate agent, but generally, how do you think about the role of an agent versus the role of tech?

Speaker 3 (16:38):
I want to back that up a tiny bit because I want to address the role that the agent wants to have

Speaker 3 (16:46):
And the role of tech and how they work together. So when real estate agents get their license, they're not thinking about the fact that they're going to have to make a hundred calls a day or sift through 50 leads that said, "Stop calling me. " They're thinking about showing homes and helping people get home and writing contracts and working new construction and picking out granite. That's what they're thinking about. They're not thinking about working leads and making calls. So how do we get to the point, and that's really what they're better at too, is the relationships, right? They're better at the relationships. So how do we get to the point where agents are spending more time working on those relationships and less time sifting through and trying to find the people to build the relationship with? And that's how we like to look at it.

Speaker 3 (17:45):
And that's how we have looked at it for a long time, even before AI, where we were looking at, okay, the different behavioral prioritizations like tags that come in through lead providers or focus on the people that have inquired about a property, not just registered, and then figure out how to get them to the point of inquiring about a property, surfacing those opportunities that agents want to address and using AI for that to get them to that last piece. And there's two different components there and I might be jumping ahead on you, but- No, you're good. There's two different components there. There's the operational efficiencies of AI where you can take some of the noise off the agent's plate, right? So that might be paperwork, it might be sending emails or writing text messages or those kinds of things, or it might be using and incorporating a system to prioritize leads, using AI voice, AI text to warm those leads up and get them to a point.

Speaker 3 (18:46):
So there's two different sides to that where ultimately the goal is agent leverage and we've had that agent leverage in the past by showing assistants or transaction coordinators or listing managers, but it's been more human or software and now we can move into an area where we can add even more leverage by using AI.

Speaker 1 (19:10):
So that's the playbook, that's the recipe. We've been designing agent leverage initially. I'm just going back to the origin of Teams kind of and feel free. I'm going to say something and stop and then you just add on to it or characterize or whatever. We started Teams because we needed leverage. And by that, I mean a highly productive real estate agent needed leverage. It's been happening for 30 to 40 years. It's looked a lot of different ways. It's obviously become much more sophisticated and scaled over the past 10 to 15 years or so. But the whole goal was how do I put more leverage behind myself? And then we design these systems that typically mirror the type of business I built for myself that now the right fit agent can plug into this system I've designed that has leverage for the agent to do the most important agent things, which are some of the things you already described.

Speaker 1 (19:56):
And so when we look at whatever technology was 20 years ago, I don't know exactly what that looked like, but it was obviously not what we have today. Today we have machines and data sets and filters over them and interfaces on top of them that allow us to like sort and find the right people to talk to, which is kind of top line benefit and rip out the slow tasks that we don't really enjoy, which we'll call operational benefit. And would you agree ... Now I guess I'm going to ask a question. Would you agree that in general, the AI conversation is heavier toward the operational efficiency? I'm going to stop doing this, the machine's going to do this, I don't have to do this paperwork anymore. I have this thing read contracts for me and make recommendations. Of course, I'm still human in the loop on that, but versus the top line.

Speaker 3 (20:47):
Yeah, no, I would agree that that is what most people are thinking of. They're thinking of AI for writing listing descriptions or creating social media posts, those kinds of things. If this- Email nurture

Speaker 1 (20:59):
Sequences.

Speaker 3 (21:00):
Yeah. Little nurture sequences, writing emails, thinking of if then statements like, if this field is filled out on this contract, then do this thing. It's basically just a slightly more sophisticated automation really that people are incorporating AI into, but there is a lot more that can be done there. And there's ... I mean, we're incorporating some of it and there's still a ton more that's coming out down the pipe that, like I said, it's changing every single day. It's hard to keep up with.

Speaker 1 (21:30):
Get really specific for just a minute. When you think about follow-up boss and you think about maybe that layer that Zillow Pro has added, I want to bring to life this idea of I'm helping my agents talk to the right people a lot more often so that A, morale is higher, they're less discouraged by ... Think about the last time you made three hours of cold calls and what that experience was like.

Speaker 3 (21:54):
Yeah. It takes a certain type of person.

Speaker 1 (21:55):
Yeah. How do you avoid that? Whether it's Follow Boss with Zillow Pro or whether it's other tools in your tech stack as well, how are you finding value for yourself and/or what are you getting great feedback on from either your own agents or from other agents that you interact with in other markets as well? What is really helping people reach the right people more often?

Speaker 3 (22:18):
Yeah. So it's definitely surfacing the behaviors, right? So we focus on speed to lead a lot of times and we ignore speed to action. If you've ever seen me speak, I constantly joke about the Nurture Black Hole in everyone's database and it's finding those opportunities within that quote unquote nurture black hole and surfacing them. Zillow Pro has been instrumental in just surfacing those behaviors. We spend so much time trying to get our leads to look at properties on our own website. So we're trying to change their behaviors so that we know when to reach out to them when they're still looking on Zillow and we all know that. And Zillow Pro gives us those insights and allows us to then run surface automations behind them. If they're looking into a new area, surfacing that opportunity, if they're likely to move, and that's using all of the Zillow backed data on when are they likely to move, when are they a high intent buyer?

Speaker 3 (23:18):
We're testing out the likely to sell feature right now, which has been really interesting. So pulling all of that data together and surfacing those behaviors and those insights at the right time, there is still a lot of configuration that has to be done behind that on making sure that you have the right smart list, but some of the things that we're working on now are making it to where that is done by AI. Your database is being analyzed by AI and the opportunities are being surfaced for you and that's going to change things drastically. So you don't have to get in there and run some kind of calculation on how many times this tag is added and then in 30 days or less, this person is two times more likely to buy than that person. So it's surfacing those priorities and changing them to show you a bite size piece from your database at any given moment in time.

Speaker 3 (24:16):
And that's based off of the entire collection and not just the behaviors that are coming from Zillow Pro, but taking all of those behaviors and all of those insights and surfacing the missed opportunities or the potential missed opportunities before they become missed so that agents can work those properly. I'm not 100% sure if I answered your question, Ethan. I might've gone on a tangent.

Speaker 1 (24:38):
That's okay. We're just having a conversation.

Speaker 3 (24:41):
So what happens?

Speaker 1 (24:42):
Yeah. So that's the intelligence in artificial intelligence. It's pattern recognition. And now we deploy it on your database and the way that it's behaved, the way that actions have happened, the tags, as you've mentioned, that get put on manually or by moving through a particular process in a particular way and that pattern recognition is happening. Whereas before, you, Tiffany, as the team leader, had to come up with a really good question and go look for the answer.

Speaker 3 (25:10):
Yes. Yeah. So I think this is a great opportunity to kind of introduce the role of generative AI versus agentic AI, right? Yeah, good call. Yeah, please. So imagine, so we're very used to generative AI at this point in time. Generative AI is where you put in a prompt and you get an answer, right? Write a listing description for this house with these features, and here is your answer. This is your listing description. Write an email campaign that includes 12 emails for this type of buyer or this type of seller, and it spits out those things. You put them in as templates and it's good to go.

Speaker 3 (25:45):
Then we kind of augmented that a little bit and added some intelligence layers and we could kind of do some customization per the lead and their activity and their timeline. We introduced that with smart messages, right? But now I want you to think about agentic AI, which is the next level of that. And this is where instead of saying, "I want you to do this thing, AI," we're saying, "I want you to think like a lead manager." You are a lead manager for a database. This database contains opportunities and opportunities may be surface like this or they might be surfaced like that. And there's a model that's built behind it with guardrails and it's like AI has now become a role in your business with skillsets versus you having to think of what do you need to tell it to do. You're telling it to act like this person that you would hire to manage your database.

Speaker 3 (26:43):
That's the difference between this generative product where we've had to say, "This is what I want you to do and we have to know what to tell it that we want it to do to this agentic role where now we're just training it. " Imagine if you were training a VA to go in and scrub your leads for you. Well, now we're training AI to do that for you and to surface those opportunities. It's a totally different world that we're operating under versus having to know what to tell it to do.

Speaker 1 (27:12):
This is just a natural extension. When we think about when, where, why, and how to deploy any of this stuff, the goal is to put leverage behind our agents, to help them be their best selves, whatever you think they should be doing 80% of the day, every day, support the rest of it. We used to do that with people, whether it was VAs or other staff, admin, whatever, in- house or halfway across the country or halfway

Speaker 2 (27:36):
Around the

Speaker 1 (27:36):
World. Then of course, software has been along for a lot of that ride. And then it was humans using software, and then it was AI embedded in our software, and now it's essentially agents operating across our software.

Speaker 3 (27:51):
And when you say agents, you mean AI agents?

Speaker 1 (27:53):
Yes.

Speaker 3 (27:54):
Not real estate agents.

Speaker 1 (27:55):
Right. Yeah. It

Speaker 3 (27:56):
Gets

Speaker 1 (27:56):
Awkward in real

Speaker 3 (27:56):
Estate context. It does. It does a little bit. But yes, it's those AI agents operating across our software. And the best part is think about managing a team of VAs. You still have to go behind them and make sure that they're doing what they're supposed to be doing, that they didn't miss anything and that everything is operating properly. And yes, we still need to do that with AI where we have those guardrails or judges in place to say, "This is what's happening and it's happening correctly every single time." But the beauty of it is when you do identify those pieces and parts that maybe aren't working 100% correctly, all you have to do is tweak your model and it's fixed forever. So it's not a constant management issue. It's an ongoing evolution of just improving your model. And we're doing that for you in some respects at follow-up boss with some of the product iterations that we're working on.

Speaker 3 (28:49):
There's some exciting stuff coming down that we're going to start testing soon, so I can't talk too much about it, but where we're going to be surfacing these opportunities for you and iterating on finding the best opportunities for you. And it's not just based off of one person's database or one person's algorithm, it's based off of the collective. Super interesting. And when you have information from across the gamut of real estate teams and Zillow and who ... I don't even know where all the data's coming from. I'm not even going to pretend like I do, okay? But what I know is that we're looking at this in a very granular way to find those best opportunities and to prioritize those and surface those in a way that agents want to work them. So that's the goal and that's also what's going to help that consistency piece.

Speaker 1 (29:41):
Yeah. And in a way then that helps your playbook that you've developed become a better playbook period. Two kind of high level ideas that we definitely went by. I just want to highlight them. One, when you were talking about training an AI agent like training a person It's the same thing. They need supervision. They need guidance out of the gate. They need to know what the job is. They need some context for what they're doing. We need to teach and invest in them, and we need to stay alongside them. It's not like you spend three to six weeks getting a VA up to speed on the four main projects that you have them running. You still do one-on-ones with them periodically. You still check in with them. You still make sure that they're doing the right stuff in the right way. And this is the human in the loop piece.

Speaker 1 (30:28):
When we talk about AI slop, we're talking about two things in my opinion. Common term, if you're not familiar with it, just search AI slop and you'll see a lot of funny and terrible and sad, horrible things. It's either A, using the first thing that you have generated with generative AI and just publishing it and using it or whatever. It's like saying, "Write me a seven email sequence to follow up with someone I met at an open house, but who never returned any of my phone calls." And whatever it gives you on the first try, you just publish it and start running it against people. That's sloppy. So the human needs to be involved in that. The other thing is this kind of set it and forget it in an agent context where we set it up and we just think like, "Well, we gave it some guardrails and now we're just going to let it do its thing." I dare you to let something go for nine months and look back and see if it was treating people internally or externally the way that you intended.

Speaker 1 (31:20):
And getting the job done at the standard you intended. I guess that's another thing I would introduce because we talk about it all the time on the show. Standards and accountability. The human in the loop needs to set the standard with the AI agent and monitor it and hold it accountable in some way. It's obviously different than holding a human being accountable, but human in the loop is one. The other big category, and you referenced it in talking about not just learning from your own limited world, but learning from the collective follow-up boss world, let's say, and then improving in that way. The quality of your AI is only as good as the quality of your data.

Speaker 3 (31:55):
Yes. No, 100%.

Speaker 1 (31:58):
So anyway, I said a lot there. So

Speaker 3 (32:00):
Anything come to

Speaker 1 (32:00):
Mind for you?

Speaker 3 (32:01):
Okay. So I think that's a really important thing to talk about. So I mean, when you talk about the AI slot piece, most of the time when that comes out, when those funny emails or the funny drip campaigns are listing descriptions and everybody's like, "That sounds like every other listening description, you didn't edit it. " There's a way that you can build a model around that to where you have a unique voice and then it is plug and play, but you have to build that model. So you have to say, "Here's how I ... " Instead of just saying, "Give me a seven email drip campaign for open house," you say, "Here's several samples of emails that I have written to send to people that came to an open house, now use these examples and the context that I'm giving you to then generate these emails." So that's how you train it.

Speaker 1 (32:52):
Just like you would with a marketing coordinator you hired.

Speaker 3 (32:54):
Just like you would with them. You would say, "I never say, dear Ethan. I say hi, Ethan. You're training the voice." So the same concept there. And that would be the case for AI for a lead manager, any of those kinds of things. It's only as good as the data that it has. So I mean, one thing I can tell you from a follow-up boss perspective, smart messages and follow-up boss, which if you haven't tried to use them, please do because it's really very efficient. We have seen such an increase in messages sent. Our response rates are better. We're getting through more leads, all of those kinds of things. It's better than just sending a text template. So instead of saying, "I'm just checking in on you, " you're saying, "I'm checking in on your home search in Chesapeake, Virginia. I see that you were looking for a four bedroom home six months ago.

Speaker 3 (33:48):
Has your timeline changed?" That's context that's relevant to that human being and it's built off of the information that is held in the profile of that person in follow-up boss. So the more data that you have in your system that is feeding your model, the better your output is going to be. And that's going to affect the lead prioritization piece as well. It's not just about creating those smart messages that tie in a location in a number of bedrooms. It's about creating this data model that then is going to be able to help surface opportunities for you better in the future. So I mean, if I can tell you anything right now that you should be doing to prepare for the future of software and AI and how it is going to service your business, it is make sure everything is in your CRM. And for follow-up boss specifically, make sure you're using your dang phone number.

Speaker 3 (34:45):
Use your follow-up boss phone number. Oh my gosh.

Speaker 1 (34:48):
But tell us why.

Speaker 3 (34:50):
Having those call transcriptions in there is just going to help, again, surface those opportunities. It's feeding that data model. Add notes after listing appointments. We have a feature now where you can talk to your phone and add a note, and we're actually working on improving that in the future. But even right now, you can hit the button and just transcribe your note. All of that is going to feed in. So think about if you go and you show a home and the insights that you have from that meeting, they don't make it into your CRM most of the time. We keep it in our head and we think, "Oh, well, we know that they don't want to have a home with a pool because their dog won't stay out of it, " or whatever it is. But if you can add those notes into the CRM, as we iterate and as our AI gets more intelligence, more intelligent, when we get to the point where we can start looking and finding properties on behalf of the buyer, and some of these things are being looked at, are being worked on.

Speaker 3 (35:52):
When we get to the point where we can start doing that, if those notes aren't perfect, aren't in the system, then we're not going to eliminate the homes with a pool. So start getting in the habit of getting everything into that system because yes, AI is only as smart as the data behind it.

Speaker 1 (36:12):
I love that. And again, this is feeding when you click to send a text message to someone that has some default settings on it in a smart message, it's pulling from all of that stuff to craft for you in a fraction of a second, a very well crafted text message that you can eyeball and change if you want. It's also learning from the edits, from my understanding.

Speaker 2 (36:31):
Yeah.

Speaker 1 (36:32):
It's taking an email or a text message that it's smart generating for you. And if you change it the same way three times in a row, it'll stop doing that essentially. Yeah.

Speaker 3 (36:41):
We're working on learning the voice of the agent as well. So how that changes and how the voice of the agent changes depending on the lead. You have different people you talk to differently, right? If you've got a past client that already knows, likes and trusts you, you're going to speak to them differently than a new client. So all those kinds of things. So incorporating, again, not just how you're talking, not just the changes that you're making, but also incorporating how you're talking to specific leads. All of that factors into the model.

Speaker 1 (37:11):
Really good. And I just want to double back too on this idea of making sure you're using the fub number. And I just want to give an example from my seat on this side of the microphone, which is, I produce these conversations, I'll meet with you in advance. We kind of settle on a couple key themes. I come with an outline of what we're going to talk about. Sometimes we adhere to it, sometimes we don't.

Speaker 3 (37:31):
And then I throw you a bunch of curve balls.

Speaker 1 (37:33):
Yeah. Yeah. Well, that's it. I mean, it's a conversation, right? So I just want to make sure I know about where I want to go on behalf of the audience. But as you're having a conversation, let's just say you're sitting in your car, safely parked in a parking lot between things while you're running around in the middle of the day, someone calls you or you have five minutes, so you're going to call them. That's the situation I'll set up. You're calling from the fub number and it's taking that whole transcription and that all gets applied to that person inside your CRM. And the reason that that's helpful isn't just so that you remember three months later or six months later that it can pull those details into a smart message for you. Instead of just being in your head, I will just remind people that it doesn't always make it into your head because I still edit all of these episodes because I find it, A, a great way to revisit the conversation in a way that allows me to understand it differently.

Speaker 1 (38:25):
I'm having this conversation with you, Tiffany, but I'm also wondering, what else do I have on my conversation outline that we haven't touched that I maybe want to touch? How much time have we been recording? I want to respect the audience's time a little bit. There's a lot going on. And so if you're sitting in a parking lot and someone has a flying shopping cart coming across the parking lot and you're wondering if it's going to hit your car, you might actually miss six seconds of what someone said as you're paying more attention to what's happening outside your car than what's happening on your phone. And so these are details sometimes you miss. And so I learned things from my guest. I had the conversation. I hosted the conversation, but in watching it back, I learned things from it. And then same as we're doing with all the call transcripts.

Speaker 1 (39:08):
I have 160, 15 to 20,000 word documents now that I have 160 some episodes and I can point Claude Cowork at them and ask for themes, topics, questions, give me five quotes about this, et cetera. And it does things we can't do in a reasonable amount of time and it's pretty

Speaker 2 (39:27):
Exciting.

Speaker 1 (39:28):
What are you most excited about? Whether it's an upcoming product feature as this episode releases, Automations 2.0 will be alive for the entire world. What are you super excited about in this zone that we've been talking about, whether it's a specific tool or a feature or a benefit or whether it's a trend at a higher level?

Speaker 3 (39:47):
I'm going to answer that question as a trend, but I'm also going to say that you can turn on Automations 2.0 right now in your account. So all you have to do is hit the button to convert over. So that is live and ready to go. And there are a lot of really great new triggers that are in Automations 2.0 that can help you just add leverage to your team. Just need

Speaker 1 (40:11):
Two minutes on triggers for someone that's like, "Oh, I haven't been doing it. " Because like you, I often hear, I love follow-up boss, this turned into a big follow-up boss episode. I love follow-up boss, but I know I'm not getting enough out of it or I know that I'm not doing all the things I could be doing. And I feel like this is a category that could be an unlock for some people. Talk about what you mean with triggers in this context.

Speaker 3 (40:35):
Okay. So a trigger is like flipping the light switch. When you flip the light switch, that triggers something to happen, right? The action, and when we talk about automations, we talk about triggers and actions, and then there's stuff in between like delays and conditions, the if statement. But when you flip the light switch, the light turns on. That's a trigger. You flip the switch, then what happens next? The actions are the light turns on. Any of the other things that might happen when you flip a light switch, that's pretty much just it, the light turns on. But triggers and actions, that's what it is. So the trigger is what sets off the string of dominoes, that first domino that falls down, then everything that happens next. So you need something that is going to set off the automation and that's what a trigger is.

Speaker 1 (41:32):
Cool. And I interrupted your trend.

Speaker 3 (41:35):
My trend of what we're working on with product and the future of Followup Boss, and I see this across multiple technology companies, right? Not just follow-up boss, but we're working on efficiencies that enhance the consumer experience. So it's not just efficiencies, it's efficiencies that enhance the consumer experience. So why is that important? Because friends, when we have customers and clients that have had amazing experiences, they come back to us and that looks good on everybody. So that is the trend that is really exciting to me is it's not just about agent efficiency, it's not just about shiny tech tools, it's not just about a sparkly button that's going to help sell more seats. It's about creating and enhancing the experience that that buyer or seller is having because the agent is more efficient and has the right information at their fingertips at the right time to act on it.

Speaker 3 (42:47):
And that is the trend that I am the most excited about.

Speaker 1 (42:51):
Awesome. It makes me think about what you are sharing with regard to just smart messages, right? So if an agent is going to be consistent and do the things that they know they should be doing, they're going to be calling, texting, emailing. But if the AI can write it for you and it's going to be contextually relevant in a way that you're probably not, just out of habit, going to go read the ... I mean, we also summarized-

Speaker 3 (43:14):
It's inefficient for you to do. You

Speaker 1 (43:15):
Could actually read the summary and come up with that text message on your own, but it's going to take you five to 10 times longer minimum. And by that I mean it's going to take you 15 seconds instead of three.

Speaker 2 (43:27):
Yeah.

Speaker 1 (43:27):
But it's probably actually going to take you a minute and a half, so maybe it's a hundred times longer or something. I don't know that that math works out, but the point's the same. You're going to take a minute or two to figure out what you should say to this person to engage them in a helpful and meaningful way, whereas because the AI is quickly surfacing everything that you've talked about over the entire relationship with this person and you want to text them right now, that's already a better consumer experience. So the idea that this is the filter through which a lot more work is being done is super exciting. On Techtember, episode 76, where she and Kyle Draper both break down best practices for AI calling and AI texting, great episodes, super practical, episode 76. I didn't ask you these because of the format that I made up for that series.

Speaker 1 (44:12):
So I'm going to ask you these for the first time in like two and a half years. And the first one

Speaker 2 (44:16):
Is,

Speaker 1 (44:17):
What is your favorite team to root for besides Gal Zenith Grouper Fub, or what is the best team you've ever been a member of besides the teams you're already a member of? And it doesn't have to be a real estate team. It can be any kind of team at all.

Speaker 3 (44:29):
Favorite team to root for is my kids' soccer teams. I love it. I love being a soccer mom. I love watching them. I was not a big soccer person before my kids started playing and they are ... I am in it. I am so in it. And that is my ... I've got two sons, so it's two different teams. So sorry, I'm breaking the rules, but it's my boys' soccer team.

Speaker 1 (44:54):
Love that. How old?

Speaker 3 (44:55):
They are 10 and 12.

Speaker 1 (44:57):
Okay. So they're playing real soccer now.

Speaker 3 (44:59):
They're playing real soccer. Yes, they are. Yep. It's a real thing.

Speaker 1 (45:04):
Do they have proper jerseys? Because when I was playing soccer as a 10-year-old, it was like wear something red. I feel like

Speaker 3 (45:11):
It's

Speaker 1 (45:12):
So much more developed now.

Speaker 3 (45:14):
Yeah. So they both play in advanced leagues on travel teams. So they definitely have legit soccer jerseys with ... They've got the home and away colors and the different shorts and different socks depending on the what's going on. And they have a practice jersey and Declan was ... My older son was just invited to be in a ... I would call it the equivalent of an all- star team for baseball, but it's a combined team for their age group based off of the club that we play in. So he now even has another jersey. So yes, they have all the legit jerseys, that's for sure.

Speaker 1 (45:54):
That's cool. What is one of your most frivolous purchases or what's a cheap skate habit you hold onto, even though you probably don't need to?

Speaker 3 (46:03):
All right. I'm going to tell you my cheap skate habit. This is true confessions right now. I cut my own hair. I do. I cut my own hair and it's not really necessarily because of the money as much as it is because of the time, but there you go. That's my cheap skate habit. So if you guys see me at a conference and you notice I have a bad haircut, it's my own fault.

Speaker 1 (46:27):
You won't notice that, by the way. If you're listening on a podcast app, you could always watch them. I'd put the video on Spotify, put the video on YouTube. I embed the video when we write these up at realestateteamos.com. You can see that you would have no idea that that's the case. When did you have enough courage to ... Because there's a first time, right?

Speaker 3 (46:48):
There is. Yeah. There's a first

Speaker 1 (46:50):
Time. And you're like, "I'm going to do this. " I feel like there's probably a lot of nerves in that moment. What did you have to overcome to try it the first time?

Speaker 3 (46:58):
I mean, so the first time I ever did it was during COVID, and so I kind of didn't have an option, and then it was like, "Eh, that wasn't too bad. I can do this. I can handle it. " I mean, I have long hair, so if you screw it up too bad, you can go to a salon and they can fix it. So yeah, it's not terrible. I always make my husband or my assistant check the back and I'm like, "Make sure I don't have anything crazy looking back

Speaker 1 (47:24):
Here." That's good. Well, yeah, I mean, it's just part of the process. What does it look like for you? What do you enjoy doing when you're investing time in resting, relaxing and recharging, or what are you doing? What does it look like when you're investing time in learning, growing, and developing?

Speaker 3 (47:37):
Okay. I'm going to do my resting, charging, and relaxing. This is my consistent thing. I mean, I do love to travel, so that is one. But I will say my regular time that I use to invest in recharging is I like to walk on the beach and collect seaglass. So I am very lucky that I live very close to the beach and I happen to live close to a beach that always has a lot of seaglass. I mean, not a ton. You still have to hunt for it, but it's my zen. It's my opportunity to kind of just listen to the water. And my husband usually walks with me, so we get to talk about whatever we want to talk about or sometimes we don't talk at all and I just look for pieces of seaglass. It focuses my brain on something that is not critical and gives me just that chance to kind of recharge.

Speaker 3 (48:34):
And I don't know, maybe there's an element of grounding there. I'll have to ask Steve about that. Steve Pasinelli, there you go. Shout out to Steve.

Speaker 1 (48:43):
Yeah. I love that. I love that for a variety of reasons. And if it was everywhere, then it wouldn't be hard to do. So you do need to scan and focus. I like the

Speaker 2 (48:52):
Idea of

Speaker 1 (48:53):
Doing this solo or together, talking or not talking. That sounds wonderful. And yeah, I live on the front range of the Rockies in Colorado Springs and there are a lot of really nice walks I can take. None of them are really beach walks. Everything they call a lake here is actually a reservoir, except for the Alpine Lakes, but those are hard to get to. Tiffany, this has been a pleasure. As always, if someone has gotten to this point, they may want to connect with you, learn more about you. Where would you send people to follow up on this conversation?

Speaker 3 (49:24):
Best place is going to be Instagram. It's super easy to find me. It's @Tiffany Gelzenis and you'll probably want to look in the show notes for the spelling of my last name, or you can just Google me because I'm pretty easy to find. I'm the only one of me. It's the benefit of having a weirdo last name.

Speaker 1 (49:41):
That's great. Gelzenis.

Speaker 3 (49:43):
Gelzenis.

Speaker 1 (49:44):
G-E-L, all I's from there.

Speaker 3 (49:47):
Yep, that's

Speaker 1 (49:48):
Right. Linked up down below as our episodes five and episode 76. This was a pleasure. I appreciate you and I hope you have a great rest of your day.

Speaker 3 (49:56):
Always fun chatting with you, Ethan. Thank you.

Speaker 2 (49:59):
Thanks for checking out this episode of Team OS. Get quick insights all the time by checking out Real Estate Team OS on Instagram and on TikTok.