Data Point of View

Using Machine Learning to Provide an Exceptional Customer Experience - Simran Singh, Green Dot Financial

March 21, 2022 Mobilewalla Season 1 Episode 6
Data Point of View
Using Machine Learning to Provide an Exceptional Customer Experience - Simran Singh, Green Dot Financial
Show Notes Transcript

AI, automation, and data are the future of the business world. But, it will take time for many organizations to determine how to make the most out of them. Tech companies such as Green Dot are taking the lead, and they are an example of how machine learning and data drive business growth. In this episode of Data Point of View, our host Laurie Hood welcomes Simran Singh, Vice President, Banking as a Service at Green Dot. Simran shares details from his incredible career. Although he did his undergrad in computer science and mathematics, he didn't become a techie. Instead, he turned to business and entrepreneurship.

D0E9EC4B_8 - Mobilewalla - Data Point of View - Simran Singh

[00:00:00] Simran Singh: When I say you've got quality over quantity, we don't want like, you know, to sign up 500 million users and then, you know, offer one thing that works for them and make money off that. We would rather identify the population that will benefit the most from our services, and offer them better quality support.

[00:00:19] So a better quality features, better quality support, better quality experience, everything included. 

[00:00:49] Laurie Hood: Thank you for listening today. I'm Laurie Hood. CMO at Mobilewalla, and this is Data Point of View. Data Point of View is a podcast for anyone interested in using machine learning and consumer data to achieve business objectives. Joining me for this episode is Simran Singh.

[00:01:06] Simran is Vice President, Banking as a Service at Green Dot Corporation. Green Dot's mission is to deliver trusted, best in class money management and payment solutions to their customers and partners seamlessly connecting people to their money. Simran also spent four years at PayPal in multiple roles before has moved to Green Dot.

[00:01:32] Simran, welcome to the program. And thank you so much for joining me today.

[00:01:36] Simran Singh: The pleasure's all mine, Laurie. Looking forward to our conversation. 

[00:01:40] Laurie Hood: Well, fantastic. So we're going to go ahead and just jump right in. We're going to start talking a bit about your background and the path that led to your current role, and then focus on how you and your team are using machine learning to drive innovation and customer experience and close with your thoughts on 2022.

[00:02:00] So you've had a variety of roles in your career, in different industries. You received your undergraduate degree from the Indian Institute of Technology Delhi. And from what I hear, the Indian Institute of Technology is where all of the smartest people in India go. 

[00:02:17] Simran Singh: That's right. 

[00:02:18] Laurie Hood: I have heard that from all of my colleagues who have gone there.

[00:02:22] So there may be a little bit of a internal Mobilewalla bias on that. But then you went back super impressive and got your MBA from the Wharton School of business. So talk a little bit about how kind of your early jobs, your early roles, your education have shaped your career and your career choices.

[00:02:43] Simran Singh: Yeah.

[00:02:44] It's been a very interesting journey. And looking back at, at it, it gives you, you know, at times like this, when you take a pause and look back at it, it, it reminds you that some deliberate decisions and some spontaneous decisions have brought you where you are. So I, I did my undergrad in computer science and mathematics. And unlike

[00:03:03] everybody, who, who goes into the computer science space and becomes a techie, I decided to delve in the business side of things. I was always fascinated by the fact that you can create the best product, but if you don't know how to sell it, if you don't know how to package it, if you don't know

[00:03:17] bring it to, to, to, to, to life essentially, th there's the, that for me was the more challenging part. So after my undergrad, I went to a management consulting firm, which I think is one of the best training grounds for anybody out there. The firm was called Parthenon and I was employee number three for their emerging markets practice.

[00:03:40] Simran Singh: And you'll see a trend 

[00:03:42] of me taking decisions to move to places which are not as well known, but it gives you a great chance for learning. 

[00:03:51] Laurie Hood: And I I'm going to jump in. I would absolutely agree with you on the management consulting experience. I spent five years at Accenture out of college and will, will say that I have, have kept what I learned there and the things I saw, they're with me. It is a fantastic training ground.

[00:04:10] Simran Singh: Absolutely agree. I use some of the toolkits, some of the ways you analyze a problem, even as of today. Even though that things change over time, but the principles that you use still remain the same. So I spent two and a half years with Parthenon, primarily working in their education practice, helping both not-for-profit and for-profit education institutions,

[00:04:34] do international expansion, M&A, just thinking about this strategy overall. And one of the clients that I worked for was a private equity fund called Baring Private Equity Asia. Maybe listeners in the US might not be very familiar with them, but they are the largest Asia focused fund, when it comes to private equity funds. And they liked the work that we had done for them,

[00:04:57] and they asked if I would become a part of their investment. Very naively, I still remember this conversation, I have thought I would be a lifetimer in management consulting, "This is what I want to do for the rest of my life." And when they made the offer, I remember speaking to my partner at Parthenon. He said, "Simran, even if you want to stay, I will not let you stay because the only way you evolve is by taking risks and doing things differently than you've done before."

[00:05:25] So spent two years at, at, at Baring, learned how to analyze companies and understand that even if they're not perfect companies, you can financially engineer a way to make them a better than what they are in legal ways. And, and then, I think at the back of my mind, the entrepreneurial bug, the idea of doing something on your own was always,it was always scratching me and itching. 

[00:05:54] And with, with two other friends from IIT, Indian Institute of Technology as well, we started, you know, when we got together, we didn't have an idea. We just wanted to do something. That was all that we were trying to do. And e-commerce, or online shopping was very nascent in India. So, what we built was a company called Bite, which the easiest way to describe that would be the Google Shopping equal into India.

[00:06:20] So you have one interface and you can get access to all the different marketplaces. And, you know, just like that, we had grown to like a million dollars in ERR. These 3, 23-year-olds, 24-year-olds did not know what to do from that point onward. And we saw ourselves plateau for six months, not knowing how do we take it from. We knew how to take it from zero to one, but one to ten is a very different skillset that is required.

[00:06:51] So we decided to sell it off to a Sequoia Capitaltech company, which wanted to use our technology. So what they do is they're a publishing house, a publishing tech platform. So, whenever you're browsing, let's say, an article about Motorola and you can actually, our technology would enable them to showcase how Motorola phones are available at these different marketplaces,

[00:07:16] Simran Singh: and then they can take an affiliate commission out of it as well. So, then came to business school at, at Wharton. I think this, this experience had shown that just by, you know, just by doing things you can get somewhere, but if you know how to do things, you can go, go even further. Loved my two years at Wharton. Some of my best friends are, are still friends that I made that at that place.

[00:07:40] And then when it, when I graduated, I decided to work at a tech firm called VMware, where I was doing product management. And you will start seeing now that I've had my fingers in multiple pies, and that has been a very deliberate decision. So after VMware, I went to start the strategy in business development functions at a very small FinTech called Hyperwallet.

[00:08:03] People had never even heard of it. It was actually a Canadian company that was looking to come to the US. But my, the thing that I used to make career decisions is the work that

[00:08:14] I'm going to do, is it going to have an impact. And that's all that matters to me. It doesn't matter if it's a product related role, it's a business development, it's a strategy role.

[00:08:23] We grew Hyperwallet to a fairly large scale. And we were eventually acquired by PayPal in 2018. Did spend some time with, with PayPal for a while, and now, as you mentioned, at another FinTech called Green Dot, which, where again, we are, I think, a fairly under utilized assets and, and an under-recognized brand name as well.

[00:08:45] But kind of solutions that we provide are exactly the kinds of solutions that a large majority of the American consumer is looking for. 

[00:08:53] Laurie Hood: Well, and, and that sort of tees it up to get to my next question. Switching gears to your current role Green Dot's got really interesting evolution story. So can you talk a little bit about kind of how providing your financial services products really led you to become a best in class technology company?

[00:09:14] Simran Singh: Yeah.

[00:09:15] If you thought my own professional journey was interesting, wait till you hear about Green Dot. So Green Dot was actually started by a DJ and you heard that right, a disc jockey. You know, when he would go play at concerts and, and, and music venues, he'd realized that most of his target population is students and they didn't have access to pay by card.

[00:09:36] So if you know, he would link them to his, let's say his music set that's uploaded online and say, "Hey," which was before Apple Store, right? So, " Hey, pay using a debit card or credit card." They weren't as prevalent then, and you know, the easiest way would we to use your, your parents' debit card and then they could actually,

[00:09:56] you know, track how much spending you're doing, which was not what the students wanted. So he started a prepaid card company. Essentially you get a card from a Walmart or Walgreens and you could load funds onto that card. And, you know, grew it to be a successful business. It went public in 2009.

[00:10:14] But I think what I've always appreciated about Green Dot is that much like any other evolving player they've realized the changing of times. They realize that prepaid cards are becoming the norm. So they actually went and acquired a bank into 2011. We are one of the few, I can actually name them on, on one hand, a few FinTech players that have a banking license as of today.

[00:10:39] Simran Singh: And then, you know, as digital commerce, online marketplaces, such as Uber, Amazon started becoming more and more prevalent. We, you know, Green Dot thought about, "Hey, what can we do for us to be able to serve more customers, maybe not directly, but through a partnership?" So, one of the first companies that we tied up with was Uber.

[00:11:02] As an Uber driver, you get access to a Green Dot bank account and Green Dot card which allows you to get paid instantly. You can use that to get, get cashback rewards. You can get a discount set, say a gas station or a auto repair shop. So suddenly from this plastic prepaid card company, which is available in a retail store, we had transformed, not overnight, but over 15 years into a technology company, which was offering

[00:11:33] our technology and our banking license to partners who were looking to offer financial services. And you know, it's one now, one of our largest growing segments, when it comes to offering banking as a service to our partners. Some of our partners names that you might've heard of Apple, Uber, Amazon, Intuit, Stash, Wealthfront just the name of few. 

[00:11:56] Laurie Hood: That's awesome. And it feels like that's a growing market in the US. You know, emerging financial services companies really do need to differentiate to, to win market share. Can you talk about some of the things that you're doing at a high level with machine learning? Let's pull that in that makes you better than the competition.

[00:12:21] Simran Singh: Yeah.

[00:12:22] I think it's, you you're fairly right that financial services can become very commoditized. But when you think about the population that you're looking to serve, it's very important to understand what are they exactly looking for. What is working, what is not working? The traditional way of doing it was, "Oh, we'll do surveys and focus groups."

[00:12:43] And there's so much noise in those signals. You know, you're a technology company, so how can you use technology to gleam out the similar type of insight as, as to understand your customer preferences, what is working, what is not working? And that's what I like about the way Green Dot uses machine learning.

[00:13:02] It's not a buzzword that we throw around and say, "Hey, you know, we are a machine learning driven company." It's so easy to say, "What's the use case that you're serving using machine learning?" And for us it's to be able to understand our customer preferences, and to be able to cluster our customers as well. To understand, "Okay, this is one specific set of customers who are using

[00:13:25] Green Dot primarily for, you know, as a store of money. There's another set of customers who are using Green Dot primarily to be able to shop online. There's another set of customers who wants to be able to deposit cash at a retail store." And by being able to use machine learning to cluster, and understand the behaviors of these clusters differently,

[00:13:49] Simran Singh: I think that provides us a competitive edge because when let's say, Uber comes and talks to us, we can help them understand what exactly it is that their drivers looking from a financial services perspective. Obviously nobody knows that driver's driving preferences or timing preferences than Uber.

[00:14:09] But helping them understand the financial services aspect of it, and being able to back that up using data that we've gathered either through clustered analysis or,some other behavioral patterns that we've recognized, that is such a powerful way of building a partnership, not a, not a vendor relationship.

[00:14:27] Right? Like they, they then understand, "Okay. If we work with Green Dot, not only are we getting access to the best in class financial services, but also a partner that will work together with us to understand what is the best experience we can provide to our drivers." And that's what matters the most of them.

[00:14:43] Laurie Hood: And then how do you, how do you use those learnings to impact the customer experience on your end?

[00:14:52] Simran Singh: Yeah.

[00:14:53] Once you start analyzing those, those behavioral patterns of the clusters, three things emerge. One, it helps from understanding what features we need to build, what, what are, what are lacking, what are things that customers are just not using. And we can actually pull them out. And you can do it by different clusters rather than for the entire population.

[00:15:14] That's, that's the easiest way of understanding it. Second is the ability to offer better customer support. What I mean by that is, because you are able to understand the behavioral patterns of each of these clusters, you can actually assign customer support which works better for one cluster versus the other, do train them on more of those aspects of a certain cluster.

[00:15:37] And third is in terms of then being able to either sign newer partners or open up a financial services for a bigger group within the, within an existing partner. So with Amazon, for example, you know, let's say we are currently serving their driver population, who's financial service needs will be very different from their seller population.

[00:16:01] Simran Singh: Right? So Amazon, it's a beam that it serves so many different populations. So it's this combination of being able to understand what, what is the right feature set that works for population, what's the right customer support that works for that population. And then being able to open up more of these populations, either within an existing partner or a new partner that machine learning allows us to unlock. 

[00:16:25] Laurie Hood: Very cool. So when we were preparing for the session, you know, you shared that the play at Green Dot is more of a quality relationship with that customer versus a quantity relationship. What does that mean for your business? How do you, how do you really think about that and sort of tie that to what I said earlier, kind of the mission statement, which is, you know, Green Dot connecting people to their money?

[00:16:54] Simran Singh: Yeah. So the relationship with money is... 

[00:16:57] Laurie Hood: I wish I had more of it.

[00:16:59] Simran Singh: That's that's, that's a constant, "Oh," we hear from everyone. But think, you know, when you're a technology company, often the services that you're offering to an end-user, the responsibility that you have can vary. But when it comes to handling people's money and the kind of population that we're serving, it's mostly unbanked or underbanked who are, you are, you know, depending on Green Dot to be able to do, to run their daily lives.

[00:17:27] So what, so when I say you've got quality over quantity, we don't want like, you know, to sign up 500 million users and then, you know, offer one thing that works for them and make money off that. We would rather identify the population that will benefit the most from our services, and offer them better quality support.

[00:17:50] So a better quality features, better quality support, better quality experience, everything included. So it's, that's what I mean by quality over quantity. It's not, "Hey, we will unleash every feature that, that, that, that we can build." No. We want to understand what is it that works for a specific population,

[00:18:12] how can we package it in the best way that it works for them, and then bring it

[00:18:17] to life so that they are actually getting real value out of Green Dot and not just, you know, it's not just another financial service that they're using. 

[00:18:28] Laurie Hood: And that ties to kind of what you had said previously about using that machine learning to really dig into what your customer's needs are, what they're doing, and then how can you better serve them?

[00:18:42] Simran Singh: Yeah. And, you know, 30% of people who have a Green Dot account, it's the primary bank account, being through, through our partner or directly through Green Dot. So think about that, right? Like if you're not serving them in the right manner, and on top of it, it's their money that you're handling. You know, let's say it's your Netflix streaming service goes down for 10 minutes.

[00:19:09] Fine, it's an inconvenience, but it does... it's not impacting your life. Some people might argue otherwise, but that's at least my. But if your bank goes down, if the balance that you see in your account is not correct, if you, you know... We don't charge for overdrafts. Think about who gets hit with most with the overdrafts. People who cannot pay the overdraft fee.

[00:19:32] So those kind of a building, that kind of an experience driven by insights that we glean from machine learning, that I think is the most powerful thing you can build for a customer. 

[00:19:44] Laurie Hood: Well, and there's th there's just a level of trust with the banking relationship. 

[00:19:49] Simran Singh: Absolutely.

[00:19:49] Laurie Hood: And that's huge. You know, you're, you're in a business that's highly driven by data and AI. You know, we talked a little bit about the use of machine learning. So when you're looking at your data science and your machine learning strategies and and where you want to invest your resources, what are some of the key principles?

[00:20:10] What are you thinking of as a leader in terms of how you're driving this in the organization?

[00:20:17] Simran Singh: I think the biggest challenge for us is just the amount of data that we get. I think at least 50 million people have had a Green Dot product at, in their, in their lifetime. That's like one sixth of the US population, right? And now think about all the transactions you do with do on, on that account. Volume of the transaction, dollar value of the transaction, frequency of the transaction, where are you spending the money. So it's just, it's an overwhelming amount of data that you, you get to play with, which is very exciting from a data scientist perspective, right? So there's no, there's no dark tough quality data that you're getting. In terms of the principles that we use from a data analysis or our,you know, data intelligence perspective,

[00:21:02] it's, we don't do data analysis just for the sake of doing data analysis. There always has to be either a business problem that you're looking to solve or some form of additional insight that, that this data analysis would provide. So I don't want to know 

[00:21:20] 75% of our transactions happened between 6:00 PM and 9:00 PM.

[00:21:24] So what? Oh, what if we were able to incentivize users to spend more during 6:00 PM to 9:00 PM? Because 75% of our transactions that are happening at, at that stage. Those are the kind of insights that data business insights team is supposed to bring to the, to the leadership team. And to be able to drive our business both in terms of getting your new, new users or making sure that existing users get more value out of it.

[00:21:54] So just data analysis for the sake of data analysis is anybody can do it, and it's, it's not adding any value. 

[00:22:02] Laurie Hood: No. Very, very, very good insight. So as we head into year three of the pandemic, who would have thought we would said that two years ago, know, COVID's affected businesses in different ways, you know? How do you feel, how do you feel, how's COVID impacted your business, but how, how you're working internally, but also you know, have you seen that relationship with the customer change?

[00:22:31] Simran Singh: I think, I love to joke about the fact that people often ask who was your main, who was your chief digitization officer and COVID-19 should be the answer. Because both internally and with our customers, what I've seen is the, the rate of change got accelerated. You got a lot of the growth pulled in our front loaded,from a, from a user growth perspective or user value perspective as well.

[00:22:56] Internally we were one of the first companies to go fully remote forever. So we were originally based out of LA, 90% of our employee base was within 10 miles of the office. Oh, it's LA, so it might be, might be even higher. I'm based in San Francisco, I would have never considered an opportunity in, in Southern California.

[00:23:16] I was just talking to our head of HR the other day, and right now I think 75% of people we've hired over the last year have come from outside of LA. So suddenly you get access to such a large talent pool, but you also have to make sure as, as the leadership team that you're equipping the team with the right tools to make sure

[00:23:36] Simran Singh: they are, they're being successful in a remote environment. One of the principles that our CEO imbibes on everyone is responsibility and trust. So I should be able to trust you and you should be responsible for the outcome of, of whatever is being asked of you. So, it's, we had the, the chief people officer of Netflix present our all hands recently.

[00:23:59] And she said, "If you think you've figured out what it means to work in a remote environment, you're kidding yourself." Because it's still evolving, you know, we're still figuring what's the, what's the right things. And then from an external perspective, as you can expect for a, for a business that runs more on digital payments and even some plastic payments, the use of cash decrease tremendously.

[00:24:20] We are still, you know, we're the world's largest economy, but our cash dependency is still very, very, very high as compared to some of the other economies that you 

[00:24:28] see, China, Southeast Asia, even Western Europe. There's not as much cash usage, but over the last few years, the number of people who have started using digital payment, and very surprising that the number of people have continued to use this new payments once they've gotten the taste for it. That is something that you wouldn't have expected had two years ago when the pandemic hadn't hadn't been brought in. So in other words, what you, what would have happened five years after 2019 happened in two years after 2019.

[00:24:59] Laurie Hood: Compress that cycle. 

[00:25:01] Simran Singh: Compress the growth here. And it's, it's, I think it's, it's for the better. It gets, when you give access to the same tools that had availed the rich person has at their disposal to... that's the only way you can bring someone out of the, you know, the, the poverty cycle

[00:25:18] of the ability to improve their financial lives tremendously, not just by a few, few, few inches.

[00:25:24] So... 

[00:25:24] Laurie Hood: Yeah.

[00:25:26] Simran Singh: That, that mission it's, it's something that dictates every business decision of ours. Again, it's not meant to be just a empty line or empty slogan. We seamlessly want to connect everyone with their money and specifically want to serve the underbanked and unbanked. 

[00:25:41] Laurie Hood: That's so cool. 

[00:25:43] And, and as you said, I mean, if you, if you can't get access the tools, to capital, to credit, you don't have the ability to kind of pull yourself up and change your circumstances.

[00:25:59] Simran Singh: That's correct. And it's easy to say, "Hey, pull, pull yourselve up by bootstraps, by your own bootstraps." But if you need to provide them with the right support to be able to do that and financial support, be in terms of the tools, the credit, lines, et cetera. That's what, what will eventually make their lives better. 

[00:26:18] Laurie Hood: Yeah. And then, and it lifts everybody when we can, you know, when we can that. 

[00:26:22] Simran Singh: Yeah. 

[00:26:23] Laurie Hood: So kind of as we close and we, we sort of went in this direction, talk about, you know, what you see for financial services, 2022 and beyond. I mean, it seems like it's just become, it was, it was an area that was so an industry that was so stable for so long.

[00:26:43] And now it's just incredibly dynamic. And you've got what's going on in the US as well as what's going on in emerging markets, which you can think of even is changing more aggressively. So we'd love to hear sort of your quick thoughts on, on what you think is coming. Look in your crystal ball of financial services. 

[00:27:03] Simran Singh: I like to joke that I run the ministry of hammer, so everything looks like a nail to me. But, I, I genuinely believe that over the next 5 to 10 years companies that have done a great job establishing a direct relationship with their customers, have built trust with them. The Googles, the Amazons of the they'll play a larger role in the financial lives of their customers as well.

[00:27:24] They will use, th that doesn't necessarily mean that they would become a bank. In fact, I think they would be very averse to becoming a bank because it, it comes with its own regulatory aspects. Let's put it that way. It's not for everyone. So they would be looking for, you know, a third party partner to provide those services on their behalf.

[00:27:42] Simran Singh: But the advantage that they will have over say a traditional banking institution, is that they, they understand the behavior of their customers a lot more because they touch the different journey points and their customers a lot more than a traditional financial institution will do. So they can build products that are very tailored to providing their, to providing financial tools to their end users.

[00:28:06] And then on a very optimistic note, I think if you look at the number of unbanked and underbanked people in the US today, it is ludicrously large for an economy of our size. I think as we open up a banking services,

[00:28:20] as it becomes easier for someone to access a bank account, that number will fall very drastically over the next 5 to 10 years. So

[00:28:29] the combination of everybody who you have a direct customer relationship with can offer you financial services, and the cost of financial services goes down quite tremendously. It makes me very optimistic for the, for the financial services space moving forward. 

[00:28:43] Laurie Hood: Yeah. I think generationally as well, I look at my son who's 17 and, and the things he's doing and how he looks at financial services. And think that will have, have a big influence and there are more alternatives that are 

[00:29:00] so 

[00:29:00] much more comfortable just dealing with things in different ways using technology.

[00:29:06] You know, there's going to be... we will do another podcast a year from now. And we'll let's, we'll see what's happened as the time has gone by. But with that, Simran, I want to 

[00:29:17] Laurie Hood: thank you for joining 

[00:29:18] us, for sharing your insights and your story. I really appreciate it. And to our listeners, thank you for your time today.

[00:29:26] And please join us for another episode of Data Point of View, brought to you by Mobilewalla. 

[00:29:32] Simran Singh: Thank you, Laurie. And thank you, listeners, for listening in.