Mortgage Has an AI Problem Nobody Wants to Admit | Trish Maraski

Michael Hammond: Ladies and gentlemen, welcome to a new episode of the FinTech Hunting podcast. We have a very special guest for you today. She is a dear friend, a respected mortgage industry leader, one of the most thoughtful voices I know when it comes to innovation, product strategy, operational excellence, and the client experience. Please help me welcome Trish Baraski, Vice President of Product and Strategy at Motor. Trish, welcome to the show.

Trish Baraski: Thank you. That was such a setup. Um, I'm not sure I'm ready for all that, but thank you.

Michael Hammond: You're ready for it and you deserve it. Uh, and it is great to have you. You and I get to talk at a lot of conferences, but they're two minutes, they're five minutes here. We get to dive a lot deeper. So if you're ready, let's have some fun.

Trish Baraski: Let's do it.

Michael Hammond: I've been fortunate enough to know you both as a dear friend and an incredible industry leader. You recently reflected on a post of nearly three decades in mortgage, and you reminded people at the end of the day that the industry supports the American dream of homeownership. After all these years, what still gives this work meaning for you?

Trish Baraski: Oh, gosh. Oh, gosh, like, lay it on me right in the beginning. Um, I, you know, I'm just incredibly passionate about what we get to do, I think. So somewhere along the way, I. It just really hit me how m much everything we do is really the basis of the economy, you know, for the U.S. like, it really, we get to see it all first. Good or bad, we get to see it all first. And, you know, what we do really matters. I have kids that are in their 20s, late 20s, buying houses, like, getting ready, um, one in California. So that in and of itself is a feat, you know, for her to buy there. Um, but, like, really looking at kind of what that next generation is going to look like and, you know, kind of making it better for the people around us. So I'm not sure if I answered your question, per se, but just incredibly, incredibly passionate about, you know, what we do matters, really matters.

Michael Hammond: It most certainly does. And you and I, when we get a chance to talk, even with the cameras off and everything, we talk about the things that are really important, and we have kids that are similar age groups going through the similar things of, hey, can I afford a home? What should I be looking for? Where do I spend? Why I shouldn't just wait for rates to drop? Is now the right time? All of these things that we see in the industry, we're dealing with, uh, our kids. So let's keep going on this conversation. The mortgage industry has been talking about digital transformation and innovation for years. From where you sit, where have we made real progress and where are we still using the language of innovation without truly improving the experience?

Trish Baraski: Well, I think that we've made real progress just because of AI. It's much more accessible. You know, so what, maybe, you know, a year ago or two years ago, someone would think, I don't want to build that. That's going to take me two years. If we are, you know, a shop, a lender shop, and we're not a tech company, we don't want to build. But now with AI, it's really sped up a lot of what's possible for building, you know, so it might not be that you have to go externally to, to build a chatbot per se, or to get a chatbot, I should say. Now, it could be something that it could be, you know, assuming that's, that's where your, um, you know, kind of where your vision and goals are. But it could be something that you really think twice about potentially building internally. And that's a lot of actually what we're hearing, like a lot more kind of custom dev that's happening out there because of it. Again, there will always be a lot of buy in the buy versus build and that's appropriate, but there's really a lot more accessibility to, you know, versus just a couple of years ago.

Michael Hammond: Well, and I think to your point, you know, I think what we're now going to start seeing more of is not just the buy or the build, but really a hybrid of, uh, lenders being able to go into cursor or replit. And maybe they can't build it 100%.

Trish Baraski: Right.

Michael Hammond: But they can prototype it, they can really get it close and maybe it's 80 or 90% and then they're going to have to rely on a trusted partner, a vendor who can say, okay, we can connect all the data, we can make sure that the security protocols are there. But it's exciting that things that they could only dream up now, within a week or so they can have a working prototype.

Trish Baraski: Yeah, and Michael, even actually the exact opposite, it's never been more available for the thought, leadership and strategic thinking to be on the lender side and for them to just flex up and down with a partner like, okay, hey, you take this couple weeks and build this and then give it back to us and we'll finish it. Or, you know, know, we'll make where it all fits. It's, it's both ways for sure. So it, it's kind of exciting.

Michael Hammond: Uh, I couldn't agree more. It is an exciting time, but people got to keep leaning in. You've spoken about what it means for an organization to be AI ready.

Trish Baraski: Yes.

Michael Hammond: And that term is, is being overused. What does AI readiness actually look like for a lender or a mortgage company that wants results, not just talking points.

Trish Baraski: Yeah. So also an area that I'm incredibly passionate about. Um, I was on the lender side in ops for about 20 years. So like, so that part, you know, you and I have talked about it often. Like that part where there's, there's a demo and you know, some shiny objects and they're, you know, pitched and they're great. They're genuinely great shiny objects. They work great. However, when you implement them because the process is broken, you never really get your roi. It's incredibly painful for everyone. It's frustrating. No one really realizes why it's not working. Kind of all the things right. Um, so to me it's um. I don't even want to say it's not that hard because fixing the process can be, you know, change management is maybe one of the hardest things, you know, for anybody.

Michael Hammond: Um, um.

Trish Baraski: But not that hard being it's a simple thing like, okay, let's take a step back and let's really kind of assess what our goals are for the year. Like I, I love actually that we're talking and it's almost June because to me in June, it's the right time to take that reflection back for the first half of the year, look at the vision and goals for 20, 26 or whatever year we're in and say like, am I on track? Like, do I need to kind of reset, like, and refocus? So to me it's a great time to kind of take that pause, um, not stop working because the day to day has to happen, but just reassess and think like, okay, what's in the process that I need to fix? Like is my cost to originate there, you know, is my client experience what I need it to be, some of those high level things. And then how do I kind of fix that and then, and win with automation? Because if you try to put automation on top of it, then it just amplifies the problem.

Michael Hammond: Exactly. And as you said, as June's rapidly approaching, now's the time to course correct. I think too many times we kick the can down the road, people don't want to kind of own it. No one wants to Admit that maybe they made a bad selection or they didn't really understand the processes. One of the points that you've made that's really resonated with me is technology can be implemented in ways that does not fit real life and only makes a process more complicated. Yeah, and we talk about being on the lender side. How should lenders evaluate AI and automation opportunities so they solve genuine problems instead of creating new ones?

Trish Baraski: I think the data is incredibly important. Like, here's where, you know, you can say, or I can say, like, fix the process, fix the process, fix the process. But the data is really what tells the lender, you know, that really phenomenal reporting. And sometimes that's overlooked. It's like the quiet sleeper thing that no one really thinks about. Um, but to me, it's the data and reporting that are incredibly critical in what needs to be fixed in, you know, kind of serving up. If you've got good data and reporting it, it should serve up to you. Not just like, what are the trends? Because you probably kind of know what the trends are. Um, but it should serve up to you what your next steps are. It should really kind of direct your path to say, like, okay, we kind of assumed that there was maybe five different processes, you know, that we know there's a couple of workarounds. I'm going to make it up. There's a couple of workarounds, um, when the loan is submitted to underwriting, that people, you know, kind of timing, whatnot, depending on who picks it up, all of that good stuff. Um, but what you could see in the data and what we've seen in data is that there could be dozens and dozens, even over a hundred of different processes that happen along the way in the kind of manufacturing line, just in originations and then getting into the servicing side. So to me, it's that data and reporting that is the most critical to get you there.

Michael Hammond: And I think one of the struggles is it's one of the least glamorous things to do. In many ways, it's kind of low tech. I mean, I work with a bunch of vendors and you would think that trade show lists could be easily scrubbed and bounced off of their current list. It's usually a rush. Let's just throw it in. We got to get something out to this group. And then five years later, you look back and you're like, why do we have all of this trash? Well, garbage in, garbage out. No one owned the hygiene of the lists. And I know it's not glamorous or sexy but if that stuff's not done right, then all of the new AI stuff and trying to automate all of that is a real struggle.

Trish Baraski: I love that you said that because I totally agree that it's not a sexy thing at all. And so, like, everybody, like, I don't want to say the poor people in the BI departments, you know, across mortgage, not getting the love that they deserve because they, like, that piece is really critical for it.

Michael Hammond: Yes, it certainly is. Well, AI and automation can create tremendous efficiency, but mortgage is still, uh, built on trust, it's built on empathy. Major life decisions. Where do you believe technology should remove friction? And where must the human experience always stay and remain? Wow.

Trish Baraski: Where should it remove friction and say? Well, you know, the first and foremost, um, I think any time we're talking about a client experience, whether that's internal or external, I think we have to be incredibly mindful of keeping that personal touch. So what does that look like? Because you can have a chatbot, right? And chatbots are incredibly effective. But if it's maybe I'll give an example of like two different chatbot, you know, like two different experiences and maybe tackle it from that perspective. So the chatbot that just like, kind of is very superficial, doesn't really, you and I, like anyone, has always experienced this, right? Where it's like, no, still no. Has this helped you? Still no. Still no. Still no. And you're trying to type in like, just get me to a person. Get me to a person. Right. Speak to an agent.

Michael Hammond: Speak to an agent.

Trish Baraski: Right, Right. So, um, it's that experience versus the chatbots. Again, just in this example, the chatbots that really, actually are built on top of, you know, a really formative foundation of information that can answer the questions. There's a bunch of companies out there that have some great ones. Um, again, you know, whether, like, we've built some as well, but there's a bunch of great ones out there that really will go deeper in answering the question and serving that experience. Right? So it's. Even though it's still a chatbot and automation almost all the way through, what happens is because there's so much, um, thought leadership and personal information, if you will, not client information, personal, but personal thought process in what goes into where the chatbot is pulling from that, it is a different experience. So I don't want to say that nothing's off the table from an automation perspective, but it's really about mindfully implementing it, you know, making sure that, okay, if we implement it, are we thinking about what the client experience will be. Or are we only thinking about our roi? Because if we're only thinking. Yeah, like, if we're only thinking about the roi, we miss, you know, the. That experience has to win overall.

Michael Hammond: At, uh, Motor, you help organizations connect technology, strategy, operations, client experience. When clients come to you thinking they have a particular technology solution, what do they often discover they actually need in order to create meaningful change?

Trish Baraski: So what we like to do is really ask a lot of questions, you know, like, first and foremost. Because sometimes you think based on a demo that I saw, you know, five demos ago or whatever, like, something seemed to be good with that particular product or that particular need. Right. And so they come into the conversation and sometimes they've got it bang on. Sometimes. Sometimes it's like, yes, that's. You're exactly right on what you want. But we try to use all of our experience. We've been our, uh, senior leadership. We've been in mortgage for over 20 years, each of us over 20 years. Um, and in mortgage, like I said it, I was on the lender side for that long. So we try to bring that level of experience and just asking questions, because from our perspective, um, it never helps if we just help you build something that doesn't really help in the first place. And we're not doing our job if we're not offering you any insights, um, for what we've seen, what we've done, um, what's out in the market today, uh, so from just coming in and having the conversations, really asking questions, what's the actual output that you're looking for? Kind of taking everything aside because you might come into the conversation thinking it's AI, but really, maybe it's just this little RPA bot that actually solves. You know what I mean? Like, it doesn't have to be one thing. Let's kind of take that off the table and just be really, you know, open about what solves the problem for you in the best way.

Michael Hammond: And that discovery process is so critical, and I think too many people jump right to the end result. I need to close a deal. I need to get this transaction done. Let's move them into onboarding. And they don't really understand what pain they're trying to solve. You recently wrote about collaboration without ego. Yeah, discipline and a passion for excellence. And that vision and strategy mean little without execution. Why do so many transformation initiatives fail after the strategy is approved? And what do strong leaders do differently to make sure that doesn't happen?

Trish Baraski: And first of all, Michael, I have to say that I, uh, 100% have been guilty of that, like bringing my own ego into the equation, you know. So it was, that was learned a little bit the hard way, you know, as with many things, um, but that collaboration, like, where does it kind of fall off after it's signed off on? Um, it's usually because not everyone was brought in early enough. You know, you, you have to engage the people that are doing it. You have to engage, uh, even though it seems like you just want to get it to the finish line. Right. So I've been again in that, in that role on the lender side where I'm just trying to get something to the finish line, you know, we just have to implement, Just have to implement. It's a great thing. We all know it's a great thing. It's great thing, great thing, great thing, great thing. Um, however, when you rush through and don't include everyone in there, then first of all you're missing out on some really fantastic perspective, you know, and because I'm not magical, I don't think of everything. Like, I, I never sat in that particular seat, so how would I know that? There's that, you know, question that I should have asked. So sometimes it's about just lack of sponsorship, you know, from different departments because you didn't bring them in. And sometimes it's really about missing, uh, a step, you know, and a potential outcome, negative outcome, because you didn't ask everyone the right questions. You know, you have to go to the people that are doing it day in and day out and. Super familiar. Um, it's probably, you know, actually maybe I'll say not, it's probably, uh, it can still be a great product or thing or change, but you have to get it to the finish line with everyone and then it'll be a really successful launch.

Michael Hammond: And that's also why user adoption is so low for many of these, uh, patients, is because they didn't get buy in. And I think some leaders are very hesitant to say, well, everyone can't have a vote, stop. We're not saying everyone's signing the check and everyone, but it doesn't mean we can't get input from them, explain why we're doing what we're doing, how this is. And especially with AI, when there's so many employees who are threatened, like, am I going to lose my job? They're taking all of my knowledge. Bringing them in early explaining is so critical.

Trish Baraski: I agree.

Michael Hammond: You're someone who continues to learn and evolve. You and I talk about this all the time. What are you reading right now? And is there a book you recommend to a mortgage or fintech leader trying to navigate AI innovation, uh, and change?

Trish Baraski: So I might throw you for a curveball on the new book. Um, so my husband and I decided to. That we were going to read the same book. Now, the type of genre that he reads versus the type of genre that I read, totally different. So we had to kind of meet in the middle. And you and I are much more aligned on what I'm normally reading. Uh, so, like, I love like a radical Candor book that such a fantastic book. Like, that's typically up my alley. Uh, and my husband is the other side of the equation. So. So I actually went to my son who is, uh, in his doctorate in college, uh, and asked him because he's kind of the middle ground for reading what I read versus my husband. So, um, he recommended two books and we went with the Great Republic by Winston Churchill.

Michael Hammond: Okay.

Trish Baraski: So very different. But I have to say, um, you know, I'm a big believer also. And everything is cyclical. Not just mortgage. Everything is cyclical. And there's always something to learn, uh, from kind of what happened in the past. Um, my dad always, when I was younger and kind of forever even, you know, until he, he passed, was always like, try to learn from what I did wrong. You know, don't make your own mistakes. Try to learn from my mistakes.

Michael Hammond: Learned experiences are so critical.

Trish Baraski: They are.

Michael Hammond: People say, oh, that person's got 20 years experiences. Yeah, but are they learned experiences? Did they learn something new every time?

Trish Baraski: Right, right. So I am going with, you know, kind of. Obviously you can't always learn to your point. You can't always learn from someone else's. Sometimes you can gain some wisdom from it, uh, in a little bit of, of avoidance maybe, or just a little bit of perspective. So I'm going back to, uh, and the book. Winston Churchill's uh, book, the Great Republic was his view on what the US Kind of could and should be. So I think just really interesting for, you know, as I said earlier, the. We're the basis of the economy, you know, in our country in mortgage. So I'm really curious as I'm, uh, starting the book now. I actually just started it yesterday, so.

Michael Hammond: Awesome. Well, next show, you and I will have to. Notes and some of those things as someone.

Trish Baraski: Wait.

Michael Hammond: Devoted.

Trish Baraski: Wait.

Michael Hammond: So go ahead.

Trish Baraski: What are you reading? You have to, uh.

Michael Hammond: Right now I'm reading another AI book. It's uh, Let AI Be youe Unlocking Transformation and Mastery for sales and marketing. It's from another mortgage industry expert. So I'm reading that, and then I also supplement that with Our Faith is always important.

Trish Baraski: Yes.

Michael Hammond: So I have, um, Matt Clark's book, Hope in the Grind.

Trish Baraski: Oh.

Michael Hammond: Uh, from Churchill Mortgage. I really have enjoyed this.

Trish Baraski: Did you just recommend that to me before?

Michael Hammond: And then Pray Like a Champion is from a priest at Notre Dame that I'm reading right now. So I usually have two or three going on, but those are the three that I'm reading the most right now.

Trish Baraski: Yeah.

Michael Hammond: And as we share those, you know, for anyone who's got great suggestions, send it on, please. Here's someone who's been devoted to much of your career, to making the industry better. What gives you the greatest optimism about the future of mortgage tech and homeownership?

Trish Baraski: I think it's the people, you know. Ah. Like, first of all, like, thank you. Because I'm trying, um, whatever I can do in my bubble, you know, whatever that looks like, uh, just always leave it better than you found it. Um, but I was really struck at some of the last couple of conferences how much it's the people that make this better. You know, it's. It's the. For sure, different perspectives and different opinions. But, um, you know, even at the

Michael Hammond: dinner table, when we were sharing dinner and people were sharing different insights, and what it did is it expanded my filter. Yeah, I never thought of it that way.

Trish Baraski: Yeah. Uh, and that's what I kind of love, you know, that it's the people around that care so much about their pocket. Right. Because, like, their topic is super important. And as you said, around that table, you know, some very passionate points for each area. And it might not be an area that I'm passionate about per se, but, like, he or she will make that spot better because they cared so much about it. So to me, it's the people, because that's. We're the ones that are executing and putting all of that work, all of that AI or all of that thinking or whatever it is into practice, hands down.

Michael Hammond: And what I love, too, is people could differ on viewpoints, and it was still friendly. Everyone was respectful to one another. It was great. If somebody shared a different idea or disagreed, we could just talk and collaborate. And there were so many great things shared in that there were. As we wind down, for one, you are welcome back anytime. I love having these discussions. But as we wind down, for mortgage executives listening today who know their organizations need to innovate, but they're overwhelmed with AI, automation, competing priorities, what's one practical step you would encourage them to take this week.

Trish Baraski: Reach out to some of their peers. You know, like that's, that's. I think it just goes back to that. The people and the experiences. Um, reach out to some of your peers in the industry and see what they're doing. They might not have the answer for you, but you might get like just a nugget from each one. Uh, I think that there's something special that's in that. Iron sharpens iron always. Uh, that's one that I would say.

Michael Hammond: Well, I absolutely love it. I can't thank you enough for joining today. But more importantly, thank you for your friendship and for your thoughtfulness, the integrity and the passion that you continue to bring the industry. You have a rare ability to talk about innovation and technology while never losing sight of, of the people and purpose behind it. So, Trish, thank you for being a guest on this episode of the Fintech Hunting podcast.

Trish Baraski: Thank you.