Herman Man: AI in Fintech - Efficiency, Challenges, and Real-World Impact

This transcription is AI-generated and may contain errors or inaccuracies.
‍

Matt

My name is Matt and I will be talking with Herman Man about leading cross functional teams and adapting frameworks to fit organizational goals while addressing global team dynamics. Herman, the first question I wanted to ask you is about the CTPO / CPO concept because I'm, I'm seeing it like a lot that we are going into this like ctpo, even cpto. So like, kind of like difficult to pronounce for me. But I mean like, do you think like this trend makes sense to connect the technical role and a product role? Or do you. Or maybe it depends on the organization.

Like what is your experience here?

Herman Man

Yeah, I think it really depends a lot on the organization and I don't think think the trend will necessarily pull in one way or another. I think it really depends on who the founder is and what their strength is in. I think most founders have a product or an engineering bent in Silicon Valley, and so therefore many of the founders will want to control one discipline or another. And so having that be aggregated through one central person I don't think is likely. On the other hand, when you have a founder that is more sales and marketing driven and doesn't have expertise in that area, having someone to make that decision earlier instead of relying on the CEO is really attractive. And even though my title here at Bluevine is cpo, I do find myself wearing many hats, almost like a CPTO at times. However, I've taken on that role more so at previous companies like Xero, where I ran product and, and engineering together.

And again, the advantage there was that I didn't need to go escalate any of those decisions that needed to be made between engineering and product. I could make it much more local and quickly and streamlined through myself with, with a lot of the decisions that needed to get made. So it really depends on the organization and the expertise of, of where the founder or CEO is.

Matt

And you mentioned the previous organization, Xerox, because you have really interesting background. Before the Bluevine, you worked for the Xerox, you worked for the Microsoft. So you have various experiences on the product side and leadership side. So I'm always like trying to figure it out. What are the best things that you brought from one of those from Xerox? What are the best things that you brought from Microsoft? You learned them there and you applied on Blue Line, for instance.

Herman Man

Yeah, absolutely. I think, you know, one of the great things for working for different companies is you see a different way of solving problems and you understand the culture by which it was. It was the, the process was formed because of the people, the product as, as well as just the culture in general. And so for me, a lot of it has been making sure that we get the efficiencies and the benefits of process and of people in larger companies like Microsoft, while removing the things that aren't actually necessary. And so a practical example for, for me is Scrum or the Rice framework for prioritization. I believe in having a process that is well documented so people understand and you can go scale the organization and be more efficient. But at the same time, I'm not religious on the process itself.

I don't care in a ceremony for Scrum whether or not it's one, two or three questions or the questions rotate. It needs to work for the organization. Process for me is a means to the end and not the end. And what I've seen over my career is a lot of people let the process be the end goal. And it's not. You just got to remember exactly what you're doing. And so for me, it's that perspective of taking the best out of everything and applying it to the situation.

And for Bluevine, that really means looking at Scrum, looking at Rice slightly differently, but applying the benefits of the outputs to make sure that it works for our company.

Matt

Good that you mentioned the processes, because this is one thing that we discussed recently during our quick talk. And it got my attention. Like, you know that the philosophy and approach to building the processes and like methodologies that you mentioned, the Scrum, right, that you have, that doesn't have to be exact, like, like in a book, right? And like influences of leaders like Elon Musk, they have the influence on how do you like, evolve as a leader. And I'm just wondering, have you got like this, you know, this, this road when you, when you saw for instance, the Elon Musk, like how he's making the stuff a bit bigger, right? And maybe this, like this or other example help you to hold.

Herman Man

Yeah, I think it's really interesting. I think Elon Musk is, is obviously a really smart, talented individual and runs multiple different companies. I think for me, what I admire the most in terms of how he approaches a problem is really from first principles. And I think it's really important that you question the status quo. So what he does well is really looking at a set of processes or set of products and questions do we actually need this? And the, you know, the default answer should be we don't need it. And there should be some pain in actually explaining why you need it.

And it was really gets back down to first principles. And for me, it's very much the same way whether or not it's the process that we're running or whether or not it's the product that we're building. You know, some people call it mvp and you look at it and you're, you're like, that's not an mvp, that's way too, that's, that's not even viable. Right. And so I think the, the key thing to, to go do is really cut it down, but then also have hard questions and argue it out in terms of, you know, what are the features that we actually need to go roll out to make this a very viable product. And so for me, it's always that getting back down to first principles and then layering back up.

Matt

And regarding your industry, right? Because with the Bluevine, you make pretty modern, let's say, stock, you can call it maybe fintech, we can call it like a fintech, right? And when I look at the US market, like there's plenty of modern startups, plenty of modern products, but still when you go to the U.S. i'm European, right?

I'm going to the U.S. and I'm, and I'm hearing the stories about the checks and how hard is to get the bank account and how hard is to do the taxes. So I mean, there is so much need to rebuild the kind of software in the system that are, that are there locally. And I wanted to ask you, like, how, how do you rebuild those systems or maybe where, where do you started with the Blue Vine? What, what are the, the, like real pain in the ass for the people that you are trying to solve or the compact.

Herman Man

Yeah, it's, it's, it's a really good question, Matt. And I think, you know, there's a lot to unpack there. I think first off, you, you are right. In the US I think the, the banking infrastructure or the banking experience is, is pretty poor and largely because the systems are very antiquated. So I think the, if you take a step back, you look at, well, how did that become what it is? Well, a lot of the, the major banks today, the major financial institutions have layers upon layers upon layers of infrastructure, all the way back from COBOL and FORTRAN on mainframes, right? And so when you have a core that's running that old and you're layering on top and a lot through acquisition, really the systems don't talk well together at all.

And I can speak about that personally in my banking experience. So I bank with one of the largest Banks here in the US and they constantly ask me whether or not I want a mortgage, I have a mortgage with them. And so it basically is an example where infrastructure just, they don't talk to the systems, don't talk with one another. At Bluevine, we had an opportunity to do things differently and look at things differently, and it starts off with the customer experience. Yes, granted, Bluevine is a newer platform, and so therefore we get the benefits of building in the cloud from day one as opposed to something on prem or through a mainframe or whatnot. So we get that benefit, but knowing that benefit, how do we exploit and leverage that to actually move even faster? And so it gets back down into the philosophy and what we want.

Our core thesis for what we build at Bluevine is everything needs to be connected as one. It needs to be one central platform. And so we started roughly about 10 years ago focused on a credit problem. And what I mean by that is we actually started in invoice factoring, not small business banking. And Eyal, our founder and CEO, he had a vision that was you could leverage the Internet and leverage machine learning to actually make invoice factoring quicker and lending quicker. And so we built around that, but we always built around the assumption that that would be a building block to the larger platform.

And so that's what we've done. And so we were able to go launch a small business checking account, which again, our thesis is if you're going to have a relationship with a small business, you need to have a checking account because lending is very transactional checking. And banking is where small businesses build a relationship with you and you earn their trust. And so we built a small business banking account and we have that now. We have the ability for them to get credit so they can get a line of credit, they can get a revolving credit card with us as well. And we've built an AP and accounts payable bill pay platform from the ground up as well that is fully integrated within Bluevine. We have other products that we're about to release later on this year and early next year.

I can't talk about them yet, but it's basically our strategy where you kind of see it with a hub and spoke model, where our hub is really the business checking account. And then you have all these different products that integrate very well within them. And really the way that I think about it is it's almost like a financial OS, if you will.

Matt

It reminds me like kind of HubSpot, right? If you're using HubSpot, you have sales, marketing teams, you have customer support teams, team. So you had all the possible tools, like even the small one that helps you with the little things. So I kind of see it as this.

Herman Man

Exactly.

Matt

You mentioned one of the co founders and when I checked the Bluevine, it's a company from Israel. Right. This is where all the started, all that started. And I'm always interested in the cultural differences between like building the product or you know, dealing with the team, managing the team which is based in, in like a different time zone or even like within like a different culture, which is sometimes difficult. Right. So you are, and when, when you, when you look at it like when you look at the big picture, you're a cpo, so you have different people. Like how, like you approach the guys in us how do you work?

Do you have like your approach or how you should approach the guys? Israel, how do you like, what are the cultural differences in working here?

Herman Man

Yeah, yeah, it's a really good question. It's, it's definitely there are cultural differences or time zone differences that you just need to manage through. In terms of work week, for example, within Israel, there is the workweek that starts from their Sunday through their Thursday as opposed to Monday through Friday here in the US and so when you add that all up, you end up losing more hours. I think culturally Israel is, is really a really driven country and talent is immense there. I think, you know, I saw some capital or I saw some, some study where there was 1 in 1 7K citizens were, were startups. Right. And so it's a startup nation and they really, really hustle.

I think that's what I found with Israel when working with them, they really do hustle. They're also very direct. And so the culture there is to be to really confront the situation as opposed to being really nice with your positioning and so forth, which in the US were more likely on the latter side. And so to bring people along, it's a very different way of thinking. I think in Israel it's more about just telling the people the why, but you can tell them and be very blunt with them where in the US I think it's more convincing and bringing people along. And so there are cultural differences. But for me, really the most important thing as I'm building a team is to be able to set the norms and set the guidelines of the team.

So what I don't want to do is tipty tip, tippy toe around different offices and different cultures. I want to be very explicit in terms of, here's how I Operate and I will operate and be direct with my US team, but I'll always be respectful and likewise with Israel. And just because I'm direct with people in Israel, it doesn't mean I'm disrespectful to them. And so there's a set of tenants that I operate by and that's the only way that it'll work for me to be consistent and it'll allow me to scale as well.

Matt

What is your team structure and who is your partner in crime here?

Herman Man

Yeah, so the team structure here is I run product engineering or, sorry, I run product and design. And Nir Klar, who's a co founder, runs all of engineering, essentially R and D, and that is my partner in crime. So I work closely with nearby across our, our sprints, across our quarterly plans and so forth, and our yearly plans to deliver what we need to deliver.

Matt

I'm always interested. I mean, like I have a few friends who work as a CPOs, and when I got some conversation with, with one of them, I see there is like a strong push or strong emphasizes on the business. Right. Which is which. Which should be obvious, but, but it's not so obvious in every, every company or every startup. And I'm just wondering in your pace, right, so you have the cto like the co founder who, taking the, the engineering part, you're taking the CPO role. So do you work for instance with the business or the sales a lot.

They are getting the, the client's feedback and then that you are like making the kind of vets on kind. What kind of features do we need? Features? Do we need ROI on it?

Like, how do you, how do you approach it? Maybe do you measure the potential ROI in the future?

In the future?

Herman Man

Yeah, yeah, no, definitely. I think a really good question on pack it in two ways. I think one question was how do we prioritize implicitly, I think what you're asking there. And then the other essentially is.

Matt

Who.

Herman Man

Do I work with to get that information really? And so maybe the best way to think about this is from a prioritization perspective. We do take multiple inputs from multiple stakeholders. And the stakeholders being sales, like you said, support being another. We talk to customers as well. So we do qualitative interviews with them. We get quant surveys as well.

And of course, another stakeholder are the machines. They tell us specifically how users are using our product and so how we can improve them. But it's really through the conversations of sales or through the conversations we've had with customers where we get Insights into exactly what we think customers actually want, what their problems are, and we think how we can go solve it differently. The way that we actually approach problem solving in general is we have a thesis which is a five year plan. And the question is, is at five years, what do we think Bluevine is? And what do we think Bluevine has really done and served for our customers? We then back that into a yearly plan which starts at the calendar year.

So our fiscal year and our calendar years align on January 1st. And so I'm in the process right now of planning for all of 2025. And the planning there really is more around at the end of 2025. What do we think we're going to get done as it pertains to our five year plan? And then we have quarterly planning processes that help drive that execution of a plan. And of course the plan changes over a year. And so the assumption that we've built over the year is one where we have a high conviction over it, but we're not dogmatic about it.

And it's not that it'll never change. And so that's, that's our approach. And in that process we learn. Right. And so through the quarterly planning cycle, we're having more conversations with customers, we've launched certain products, we're seeing the feedback and we have to iterate. We'll, we'll have assumptions where we're like, hey, we thought that after Q1 we would have been good here, didn't, wouldn't need to touch us, only to find that, you know what, we were totally off here and we might have to pivot. And so we'll do that through a quarterly planning process.

But it starts with having a thesis and a target, if you will, over five years, backed up into that next year so that we're, we're, we're in the details.

Matt

And speaking of 2025, because like you, you, let's say you have the plan for five years, right? And the last two years, I mean they are insane. Insane, right. So when I think, when I'm thinking about the planning right, in the future. So the one thing, it's like the big text, like the recession, like not really planned. And the second thing, it's the AI. So I assume like in your product, AI and you are the product where you have a lot of data, could have a lot of impact.

So I'm just wondering, how do you see the, the future? How do you see 2025, 2020 26?

Did it affect your plans?

Herman Man

Yeah, there's, there's a lot. I mean, yes, absolutely. You know, we, we have a hypothesis on the next year, the next two years. Obviously there's macro conditions like the economy today is election day, so, or the post election day. And, and so there's, there's implications to that that we just don't know until, until the new term takes takes effect. And so there's only so much we can plan, but of course we make assumptions on that planning. Now you talk about AI and AI is an interesting one.

It's obviously, you know, with, with Gen AI and with ChatGPT, it's really garnered so much interest. And the way I think about it is that there's, there's AI used across a number of different scenarios and there's three in my mind that pop into, into my head. The first is really around productivity, internal productivity. Can employees use it to be more productive? And the answer is yes, we're using it across product, across engineering, across HR, across marketing, across legal, using ChatGPT for research and for generation of content materials. Right. Doesn't replace humans, but augments the efficiency and the scale.

So that's scenario one, productivity. The second one really is around proven use cases, is what I'm saying. And so proven use cases for bluevine, we use it specifically around underwriting, around machine learning, specifically around models so that we can better underwrite our customers, whether it be on our lending product or on our checking product. And then there's also the proven use case of customer support, for example, where customers are used to talking to a chatbot to get an answer. And so that makes our customer support people more efficient. Where I think it's yet to be seen, at least in the fintech world, is other scenarios. Right. Can you use it for forecasting?

I think you can, but are you worried about hallucinations? Yes, you are. Right, so you want to be a little more thoughtful there. I think the key thing again though is when we look at AI as an example, to really think about AI as a means to the end and not the end. Our customers won't use bluevine because we use AI. Our customers will use bluevine because we happen to solve a business need that they really want. And that business need could be insight into income and expenses where we could give them really accurate insight 3, 6, 12 months down the road.

But it's not because of AI that they care about, it's because of the insight that we give them. And so for me, it's really keeping the eye on the prize, which is how do we solve it? For our customers.

Matt

I'm always like to ask the question about like the contrarian approaches or, or even you could say controversial, controversial decision making. And I'm wondering do you recall any of those like in your career it doesn't have to be from the bovine, but maybe like some controversial or contrarian decision that you have made on the product level or engineering level too and you are a part of it and it worked out well. But from the outside it's not so obvious that you know like for everybody but you felt that this is the right way to do it.

Herman Man

Yeah, absolutely. I think, you know, I, I comes to two that, that come to mind. The first one was when I was at zero. We I, I ran the people portfolio and part of that was the different payroll products. So UK payroll, Australian payroll and New Zealand payroll and US payroll. We were just understaffed on the US payroll side of the house. And when you look at payroll, it's super complicated around regulations, around requirements, regulations change, they get updated on an annual basis.

You have to be able to go react. There's also compliance that's related to it.

We were just understaffed. So I made the hard decision to actually sunset the product and move on to partner with another leaving payroll provider in the US at that time. And so it was a really hard decision. I mean humans were impacted here as part of it, namely the team. There was a board conversation that needed to be had. So it was, it was a real, real difficult conversation but ultimately the right one for stakeholders as well as for shareholders as well. So proud of that decision making wasn't easy, but did that, I think an interesting challenge, you know, here at Bluevine was really pivoting us.

I talked about this earlier Matt, where we were a one product company in a lending product. We've now pivoted to a multi product suite where we have a hub and spoke spoke model with a small business checking account at its core and center. And moving from a one product company to a multi product company is a challenge. It's a mindset change. It's a resourcing allocation. It's not like we're Microsoft and we can go just hire more engineers through the whole process and more product managers. It's prioritizing, thinking about things differently, you know, getting back to first principles like we talked about earlier with Elon Musk and how he approaches things.

So those are the two that come to mind.

Matt

This is really interesting what you mentioned. Like you have, and we already discussed it, that you have plenty of small products Right, but how do you structurize the teams around it? Do you have small teams with somebody responsible for the one product or how do you approach it?

Herman Man

Yeah, on the product side I do. So I have, roughly speaking, I have it broken down into four teams. I got one team that's focused on small business banking. I got one team that is focused on our credit products. I got one team that is focused on platforms. So the interconnectivity between the whole systems and all the common objects. I got a platform team and then I have a design team.

So the design team is centralized, but the designers are dotted line into the specific teams as they're specialized in working on those said program products.

Matt

And do you set some, some like okrs or some kind of KPIs around those smaller products? Let's say to measure if they are successful, if the user are using it? If. Yes, I assume yes. Like what, what do you measure? What is important to you?

Herman Man

Yeah, absolutely. Yes. So the short answer is yes. We measure a number of different things. It basically has to ladder up to, to the company okrs. So if the company okrs have something around arpu then and contribution, then my OKR is down below. For each team has something specifically around acquisition costs.

Have something related to fraud, for example, have something related to the onboarding funnel and the conversion. So it's the product experience that we build, it's the tools that we enable for risk to be able to do their work to minimize fraud that helps with contribution there and so forth. So the way that I think about it is having very explicit and clear OKRs for my team that ladder back up into the master okrs or the company okrs, if you will.

Matt

And last but not least, the question that they asking all of my guests, it's about the recommendations. So can you recommend any like books, resources, conferences that have been particularly helpful to you as a, as a leader, as a cpo that you had this maybe aha moment after, you know, reading one of those.

Herman Man

Yeah, yeah. I mean I think the first one that comes to mind is an oldie but a goodie.

I love it. It's a hard thing about hard things from Ben Horowitz. Right. So that one stuck in my mind. There are multiple different things that, that I remember, one of which basically, and I'm paraphrasing because I don't want to swear on your podcast, but you know, one of the things he said was if you, if you eat crap, don't nibble. Right. And so I think that one is a very Physical and.

And it's one that resonates with me and I. I'm reminded every time we're in a hard situation, just pull the band aid off and deal with it. Right. So I think that's one. The other one that, that I recently started reading was, or I just finished was May Contain Lies, which is a pretty good book. It's really talking about how you can use data to manipulate your view. And so it's just a good reminder, we all know this just, just to remember to question the assumptions on the data and not fall in the trap of overly believing what you're seeing. And then of course the other Walter Isaacson books are always fun to read whether it be Elon Musk or whether it be Steve Jobs or Albert Einstein.

They're always fun reads for me just to just de stress if you will and get his perspective on those individuals.

Matt

I think the second one is really important thing that not everybody is aware about. Like the data. You can always choose the right data and where right charts to kind of support yourself with the decision. And I think that those top consulting companies are really good at doing it. So depends on kind of hypothesis they are having. They are always finding the right data to support it.

Herman Man

Exactly. And so when you read those set reports, always question and make sure that the data that they're getting isn't skewed in one way shape or another.

Matt

Herman, thank you so much for all the insights and really cool resonance. I really appreciate it. Thank you for today's talk.

Herman Man

You bet. Thanks a lot, Matt. It was fun. Follow Matt on link.

‍