In this episode of The Growth Studio, we sit down with Shru Troup, Senior Data Scientist at TIFIN AG, to explore the transformative role of AI in wealth management software. Shru shares her journey into data science, the inspiration behind her work, and how data science expertise is driving innovation in the industry.
From the challenges of integrating AI into existing workflows to real-world success stories that highlight its potential, this conversation delves into the power of data-driven insights to enhance advisor-client relationships and achieve better financial outcomes. Whether you’re curious about the future of AI or looking for wealth management software strategies to grow your business, this episode offers a wealth of insights.
Tune in and let’s grow together.
Â
Samantha Davison:
Hello and welcome to the Growth Studio. I’m your host, Samantha Davison. On the Growth Studio, you’ll hear from TIFIN AG leaders, partners, industry experts, and clients who will share their successes and the transformative role AI has played in their journey. Together we’ll uncover actionable insights, strategies and the big ideas driving growth in the wealth management space. Whether you’re curious about adopting AI, interested in hearing real world stories from firms like yours or just looking to stay ahead of the curve, the Growth Studio is your go-to source for innovation and inspiration. Let’s grow together. Today’s focus is about data science. It’s really the backbone of what we do here at TIFIN. My guest is True Troop and True is the senior data scientist at TIFIN AG. True. Thanks so much for being here.
Shru Troup:
Thank you, Sam for having me. It’s exciting to be here.
Samantha Davison:
So let’s begin by just getting to know you a little bit. What drew you to data science and how did you end up here at TIFIN AG?
Shru Troup:
Yeah, good question. So I’m someone who likes to validate. I would say my decisions, all my decisions based on data world is super chaotic. So data can become an anchor to all your decision making. They might not always be the right decisions, but it feels good to anchor your decisions on data. I’d say I spent a great deal of time my career writing code and automating systems and analyzing data to solve business problems. And so it was sort of a natural progression towards data science because it provided me with even better and sophisticated tools and techniques to find meaningful patterns in data, which I didn’t have before. And how did I end up at D? That’s been a meandering journey. I’ve worked at large companies in my early career, and after moving to Colorado about 10 years ago, my focus shifted to working for smaller, more nimble startups such as TIFIN, mainly because I love building moving faster and just being closer to business. We all know that large companies move at a snail pace and there’s a lot of red tape and bureaucracy. So being at smaller companies provide me that space where I can make an impact faster and see my work close up to the business.
And I’d say I’m very interested in personal finance and investing as well. So my own personal interest lines up with the mission of TIFIN really well.
Samantha Davison:
So then what excites you most then about working at the intersection, really of AI and wealth management?
Shru Troup:
So I grew up in India and I saw my parents super dedicated in saving every penny to build their business and their investments. And so their own hard work and their journey taught me the value of financial planning, saving money, and just that financial literacy they imparted on me from an early age. And that’s shaped me today. And now that I’m a parent to a 5-year-old, I’m very passionate about passing on the same financial lessons to my kid, especially the knowledge about power of compounding. We know that really adds on. And so what I would say is what excites me most about working in AI and wealth management is this power of AI to democratize access to financial tools, financial education advisor. We know we have seen a lot of technology improvements in the last 10 to 15 years, and we have seen what we thought about wealth management, that it was a privilege that was reserved for a few people now that technology advancement has changed that access. And I feel that with AI now, it would personalize financial planning even more and it could be offered to people who were not offered those kind of services before. So I do think that AI will continue to break down those barriers, simplify some of that complex financial decision-making, and provide personalized insights and advice to empower people like me who live in US or people in India, any individual to reach their financial goals. And so that’s been the most inspiring thing for me to be here today.
Samantha Davison:
And especially with chip. True, I mean, we say we are creating better wealth outcomes for more people, and that’s kind of the motto for TIFIN. So I think for many people there is a mystery around ai. So let’s better understand the role of data science because that really is the foundation. How does TIFIN AG use data science to drive those outcomes for wealth management firms, advisors, and ultimately the investor, the end investor, the client?
Shru Troup:
Yeah, great question. So before we talk about the data science, let’s talk about the high level challenges that the wealth management firms face. There’s a challenge around scalable ways of achieving this organic growth. And when I say organic growth, I mean not through this market growth that we have seen in the last 10 to 15 years. It’s really through how do they acquire new clients and expand relationship with their existing clients and how do they retain their existing client base? And we know just like all of us advisors have limited time in the day and a lot to do, and they have hundreds of clients to juggle and support. So I guess the question really comes down to how do they prioritize which clients to reach out to and how do they continue to deepen their relationship with their existing client base when honestly they’re spending more and more time probably forming a lot of different administrative tasks in their CRM or researching data on new prospects and setting up marketing campaigns and whatever it is that they have to do for prospecting and finding new leads. How do they find out which clients are going through different life events? So that’s a really important thing because a significant life event may signal an opportunity for a strategic wealth consolidation. So for example, without picking up and calling each and every client on their list, it is not easy for them to know which client has recently changed jobs.
I’ve been there, I’ve changed jobs in the past, and my money has sat around in different 401ks from my old jobs. And it would be nice if somebody could remind me, Hey, don’t forget to roll over your 401k and consolidate your money. And so these things are really critical and important, and as I said, advisors are spending a lot of time just in the nitty gritty and the routine tasks. So how do we get those insights in front of them so they can action, action on those insights and help clients? And this is where TIFIN AG steps in. So we are trying to solve these growth challenges. So the advisors find time to do what they do best and which is building stronger relationship with their clients, customizing plans, educating their clients, providing financial literacy, and ultimately helping their end customer realize their financial goals. That’s why we are doing what we are doing. And so at F and a G, we have built our supervised ML models to help wealth firms prioritize their clients so the advisors can spend their time on those highest ROI opportunities at the right time.
Samantha Davison:
So when you talk about building and refining models to prioritize clients, how exactly do we do that?
Shru Troup:
So there are few key steps that goes into building our models. So of course for building any model, you need data. So the first step is essentially we work with our clients on collecting that data and understanding their business use case. There is a step where we ingest the client data securely and we feed it into our data processing pipeline to clean it, standardize it, and transform it. We enhance that data then with our third party data providers to add some additional context like demographic features, financial features and things of that sort. And then after that first step, we go into the model training phase. So as I said that we use supervised ML models. So what does supervised ML really means or supervised AI?
There are two things. So what we are hearing in the market right now in the media right now is a lot about GenAI. And GenAI is essentially taking large data sets, finding hidden patterns in them, and then using those patterns to generate new content. And that new content content could be text or images, any other kind of media, whereas supervised AI, it’s a different kind of machine learning approach. And where the ML models are trained on label data sets. And what that means is these data sets include an input data. So you could have a lot of features in the model like age and income and home value, things of that sort. And then that data is paired with a ground truth or a label. And it could be a binary outcome like whether a client consolidated or churned, or it could also be predicting something like investible assets, just totally depends upon the kind of model you’re training.
So when we perform training on this data set that is labeled with this historical ground truth, what we are doing is essentially taking this ML algorithm and finding hidden patterns and relationships between these features and the output. So I guess you can think of it like creating a recipe using the ingredients you’re trying to find that relationship. So once we have created this model, then we basically take the recipe and then apply it on the unseen input data and predict the output. And that’s kind of what goes into training the model. And then there is some performance metrics that we calculate making sure our model is generalizable and performing the way we want it to perform. And one of the key, I would say, aspects of model building is then you don’t just stop. You don’t just stop building model and then never build it again or train it again.
So part of it is this feedback loop. So as the clients and advisors are taking actions, and so the activities and the outcomes of those actions are then fed right back into retraining the model. And what that ensures is that we are continually updating our knowledge of the world and making sure our model is accurate with what’s the reality right now. And so yeah, so I would say feedback loop is a really key thing of continually improving the models. So that’s kind of our steps. And once we have the model train, we score our clients, we score the advisor’s clients, deliver them these insights directly into their CRM, along with, as I said, the rationale of why somebody was scored the way they were scored. And then the clients can start acting on these high scoring top opportunities, growth opportunities essentially.
Samantha Davison:
And in that case, when you say clients, you’re talking about the advisors can then reach out to their clients. So let’s talk about the challenges that you and the team face when integrating AI into the wealth management workflows that you have in place.
Shru Troup:
Yeah. Well, of course, in order to build models, you need data. And what you really need is this unified standardized data. And a lot of the times we encounter wealth management firms that have their critical data scattered across various systems. So they might have their marketing data in HubSpot, they might have their client relationship data in their Salesforce or Wealthbox and mean, there are so many different CRMs, right? Their financial data in separate internal database. So it’s kind of just scattered everywhere. And this kind of lack of standardization makes it very challenging sometimes for both the customer and us. And our customers don’t always have a specialized data teams to do the work. So what we at TIFIN are trying is to solve this issue by using our data unification process. So we are bringing in the raw data from various sources from the client in a safe and secure manner, standardizing it, aggregating it, so we can feed into our data marts and build insights from there and of course into our models. And this process ensures that we can build higher quality models with standardized cleaned up data.
The other issue I would say is this variability in the CRM tools. So as I mentioned, the wealth management firms rely on a lot of different CRM tools like Salesforce, wealthbox, Redtail. Each of these tools have their own data model or the structure and how the data is stored. They all have different integration capabilities. Some have better, some are more robust, some are not. And they all have different user interfaces. So you can imagine the kind of challenge if we want to deliver these insights directly into each of the client CRM, we need to have a lot of engineering know-how first to integrate with every CRM out there. And so it creates a pretty big challenge for us to embed our AI driven insights into their system. So because of this, we are planning to streamline some of that integration process and enhance user usability. So our plan is to build our own standalone platform that’ll provide these insights and scores and visuals and dashboards to the advisor in a more intuitive and usable manner.
Samantha Davison:
There’s so much talk about AI that it’s going to be a game changer in the future, not now, but the truth is it is here. So let’s talk about real world examples and success stories that we have seen. Can you give us some of those where AI has made a significant impact for a client or an advisor or a wealth firm?
Shru Troup:
Yeah, so that’s the thing that’s really exciting for me, that our models are actually making a difference and making impact. We have been able to help our clients grow organically with their existing client base. So some of our learnings from our customers who have used our models indicated about 25% of highly top ranked clients that advisors engage with. They bring new money into the door, and that’s been an exciting impact to be part of.
Samantha Davison:
Yeah, I think some of the stats we throw around is, and it’s true, is that advisors who take advantage of our AI solutions are growing two and a half times faster than those who don’t. And so that’s a real result that we are seeing right now with people who are leveraging your expertise. We we’re coming to the end of our conversation, drew, if there’s one thing you would like our listeners to understand about the impact of AI and wealth management, what would that be?
Shru Troup:
So we have all heard in the news that AI is coming and it’s going to take all your jobs, but I do believe that AI is not going to replace human advisors. In fact, AI would just be another tool in their tool belt to deliver smarter, faster, and more personalized financial advice. I think it would enable the advisors to uncover insights that were previously hidden. And as I mentioned before, advisors would be able to spend less time on routine mundane tasks and more time building deeper trusted relationships with their clients and help them realize their financial goals. So be able to be part of that is exciting. So yeah.
Samantha Davison:
And it’s happening right now. Thank you so very much. I appreciate your time and your perspective, and thank you to our listeners. We hope you enjoyed our conversation with Shrew and now have a better understanding of the expertise behind our AI powered solutions. Be sure to join us in February for our next episode of the Growth Studio with more expert insights, compelling stories, and practical strategies to help you stay ahead in the ever evolving wealth management landscape. Until next time, I’m Samantha Davison and this is The Growth Studio. Thanks for listening and let’s keep growing together.
Â
Experience TIFIN AGâ„¢ today.
Tell us about your goals, and a knowledgeable representative will prepare a demo for you.