More Intelligent Tomorrow: a DataRobot Podcast

Changing the DNA of How We Serve Customers - Harveer Singh

April 01, 2022 Video Team
More Intelligent Tomorrow: a DataRobot Podcast
Changing the DNA of How We Serve Customers - Harveer Singh
Show Notes Transcript

In this episode of More Intelligent tomorrow, Global AI Evangelist, Ari Kaplan, sits down with Harveer Singh, Chief Data Architect and Head of Data Engineering and Architecture at Western Union to discuss the customer experience, blockchain and cryptocurrency, and how to digitally transform a 170 year old company.

Harveer was born in India and has traveled the world through his career - first moving to Sydney for his masters degree, then to various cities in the United States and is now in Denver. Growing up in Delhi in the 80’s, Harveer had his life uprooted and his family was forced to leave their homes.

“They were tough times. When people's lives are challenged by the politics that are played at a very high level, it's very hard to cope. I was fortunate that we didn't have any family harmed, but we lost our house. We lost everything. And we were forced to run in the streets to save our lives and for water, because we were from a minority community.”

Experiencing this hardship led Singh to strive and protect his family, while also remaining thankful of what he has been able to achieve and who he has met throughout his life. 

“If we were not forced to move away, I would have probably not tried as hard in my life that I did because I needed to make the amends that I needed to, to make sure my family is safe.”

Western Union is in the business of moving money across the globe, which can prove to be complicated. Handling money globally is also a data rich environment that Western Union has used to prevent fraud and human trafficking, all with the help of AI.

Money is also an interesting space right now, with the advent and increasing popularity of Web3 and cryptocurrency. When asked about his opinion on cryptocurrency, Harveer stated that he doesn’t currently think that blockchain is stable enough for it to be a currency used by big banks and financial institutions, but there are some benefits to the innovations around cryptocurrency. 

“​​Our financial services have become more and more complex over a period of time. I think blockchain simplifies some of those. But the underlying education is extremely important right now.”

Western Union is a 170 year old company, and while the age of a company shows stability, it also can prove to be a difficult roadblock to overcome when modernizing and building a digital product. Harveer and team were able to move Western Union’s operations to the cloud, overcoming many obstacles and building global solutions along the way. His biggest piece of advice is to lean into the existing culture of the company when making changes. 

“So some of the things that you have to think about when you deal with a company that is that old, is what is the culture of the organization? If you approach it that I'm going to change it, then you know, that is a disaster to start with. … If you approach it from that particular angle, I'm going to embrace it, but then I’m going to bring about change.”

Listen to this episode of More Intelligent Tomorrow to learn about:

  • The challenges of innovating in a 170 year old company, and how to approach digital transformation on a global stage.
  • Blockchain, cryptocurrency and the future of currency online.
  • Building customer loyalty programs that encourage stickiness and make customers come back.
  • What will shape the future, and what we may fear from advanced technology and machine intelligence.

Harveer Singh (00:01):
We want to be able to understand customer behavior. We want to understand what the customers are doing online, where they're sending money to, how much they're sending, how many times they're sending. And if there are certain occasion they're sending, why not send something else with it? Right? So it's all about how do you maximize the wallet share? But also makes sure the customer gets the most benefit out of it. We try to reward our customers with majority of the money that we receive from the affiliates. So this has to happen. You have to keep innovating.

AI Voice Over (00:34):
Welcome to More Intelligent Tomorrow. A podcast about our emerging AI driven world, critical conversations about tomorrow's technology today. On today's episode, host Ari Kaplan sits down with Harveer Singh, chief data architect and head of data engineering and architecture at Western Union.

Ari Kaplan (00:58):
Harveer, welcome.

Harveer Singh (01:00):
It's a pleasure to be talking to you today.

Ari Kaplan (01:02):
So, Western Union, people recognize the company, people recognize the logo. You guys are a leader in the global payment space, 150 million people transfer money each year, myself included. Over a half a million retail agents, 11,000 employees, one of the things that impresses me about you is I just saw you were on the cover of Digital Bulletin. That was a great photo shoot. I loved it. You looked pretty slick.

Harveer Singh (01:31):
Thanks, Ari. I would say momentarily, I was shocked that, that was me. I'm like, "These guys did a pretty good job to make me look good." But you're absolutely right. A company with 170 years of legacy and brand. It is absolutely recognizable and no matter in which country I go, I'm always smiling because I see the friendly Western Union logo everywhere I go. And it has been tested, the company has been tested and the have innovated themselves over and over again. And I think we are at the cusp of that innovation again with a lot of interesting stuff that is happening. We are changing the DNA of how we service our customers. We are not just here to help them send the money, we are here to help them with their life in many ways and possible.

Harveer Singh (02:24):
And during this conversation, I'll share some things that we have done, especially around the digital bank that we just launched in Europe, super excited. We are live in Germany and we are live in Romania. So people are transacting, which is amazing and our customers want more from us and we look forward to servicing them more and more. And with that, there's a lot of other responsibilities that comes as well, right? You need to make sure that the tech's stable, you need to make sure that we are innovating, to make sure that the risk and frauds are down. Being in financial service industry is a lot, a lot of responsibility, but we have amazing partners, we have amazing product partners, we have amazing consulting partners who we are bringing them along the journey, and it's going to be a fun ride.

Ari Kaplan (03:09):
Yeah. Fantastic. And very fascinating. And your role, chief data architect, global head of engineering. I love your background, your backstory. You've worked at Ernst & Young, IBM, Deloitte, Accenture. Why don't you tell us about what do you do now at Western Union? How did you do come to work there?

Harveer Singh (03:28):
I always believe in relationships and primarily my entire experience of consulting experience was based on relationship. My first job, when I took with Accenture was back in Australia, I was actually a professor or assistant professor. I was teaching advanced web technologies when one day my class decided not to show up. And I was like, "What happened?" And I ended up going down to see what happened and figured out that they were all sitting in a career conversation with Accenture. Accenture was presenting. So I went in the back seat, started to chat with the guy, I didn't realize he was actually the head of Accenture in Australia. And then he gives me his card and says, "If you want to work with us, come join us." And I'm like, "Okay, let's have a chat." After three months, I visited them, I was blown away by the work that they were doing.

Harveer Singh (04:15):
And that started my consulting journey. And after that honesty, I have never ever had to look for a job. It's primarily you build a relationship, when that engine works for you, it has worked for me. I had friends, I built great relationships with one organization. When they moved to another organization, they created opportunities for me to move with them to service their clients better. And funny enough, when I decided to leave consulting, it wasn't because I didn't enjoy consulting, it was because I wanted to have a sense of ownership with what I was doing. That's what I wanted to do. I wanted to build and run my own stuff and live with some of the decisions and experience that when you see, like for example, customer transacting and Western Union is doing good for people, that give you a great high.

Harveer Singh (05:02):
And funny enough before even I thought about leaving consulting and a client of mine calls me and he says, "Hey, you remember we built this 10 years ago? I want you to come and help me out building my new tech." I'm like, "Great, fantastic." And I ended up at Western Union. So right now at Western Union, I am the chief data architect and I'm head of data engineering globally. But also, it's a little loose title primarily because of the other work that I actually do. So I'm also involved in a lot of innovation work. So from building omnichannel chat bots to running customer engagements with Salesforce marketing cloud, now with owning that entire advertising technology, which is Adobe and particle amplitude tech that now my team owns. And then also helping our privacy team to become more privacy by design rather than being an afterthought. So it's more than the architecture, it's actually innovation and bringing innovative technologies to Western Union as well. So my role has significantly expanded in the let's say last year or so.

Ari Kaplan (06:05):
That's great. And yeah, we'll talk a bunch about innovation and what that means to you. Very interested to learn more how you take a 170 year old company and innovate, but talking about Australia was interesting. I know you've lived and been around the world. Where were you geographically living?

Harveer Singh (06:24):
So I was born in India and I moved to Australia when I was 22. I did my masters in Australia. I spent about the next six to eight years there. And then one fine day. I said, "I want to travel all around the world," and started my journey in New York, New Jersey area in the US in 2008. Now I am based in Denver, Colorado. And my family is based in Vancouver, Canada, which is think about me not wanting to travel, I'm on the road pretty much every other week now again, but again I'm used to it. I want my family to have the same opportunity I have been given. So I'm not depriving my better half from chasing her dreams as well. So we are trying to work as a collaborative unit, but I get the best of the board worlds. Food is fantastic in Vancouver, I would say.

Ari Kaplan (07:11):
Yes. Yeah, absolutely. And I know in some of your interviews, in our prior discussion, you talked about the touching and challenging story of growing up in Delhi and having to move away from there.

Harveer Singh (07:25):
Yeah. There were tough times, right? When people's life get challenged by the politics that are played at a very high level, it's very hard to cope up with some of those stuff. I was fortunate that we didn't have any family harmed, but we lost our house. We lost everything. And we were forced to run in the streets to save our lives and for what? Because we were from a minority community in someway or some form. And that's troubling because the political environment in India today is also bubbling up very similar at the moment. That is why it's troubling. It's very heartfelt. I'm following the news every day, but it's just troubled, right? Why can't we be and let be, right? At the end of the day? I respect everyone. I expect everyone else to respect me as well.

Harveer Singh (08:14):
I would say those times also showed me that no matter who you are, you are never safe in any very form. Unless you decide to take certain measures against it. We were forced to leave, but it gave my family a different opportunity. If we were not forced to move away, I would've probably not tried as hard in my life that I did because I needed to make the amends that I needed to make sure my family is safe, my family is well taken care of. And that's where I feel. I've been blessed that along the journey, I've built great relationships and great friends and friends like you, who we haven't even worked together, but amazing relationship that we share today, I'm super blessed and super thankful for that.

Ari Kaplan (08:59):
Yes. And thank you. And I absolutely do respect you. A personal experience I had with Western Union is one of my high school friends fortunately was on hard times and be basically was like a few days away from being homeless. And so I was able to use Western Union to send immediately. I later found out he was suicidal at the time. So it may have saved his life. So things like that, people don't realize, yes, it's money transfer, but could be very, very good, like the immediacy of it and the security of it. So any stories of what Western Union does for good? AI for good and Western Union for good?

Harveer Singh (09:42):
Many stories. You just gave me goosebumps on that one, for sure. Because very similar story, I have been a Western Union customer since 2002 when I moved to Australia because my first paycheck I sent to my parents through Western Union. I still remember standing in the line at the retail store, super excited. It was 200 bucks, but 200 bucks was a good amount of money for India. And I made sure my parents get that money. I was standing in the line and I was like, "Did you pick it up? Did you pick it up?" And I just wanted to hear their voice and how happy they were for my first paycheck. So Western Union has been embedded as part of my life since then. I did something similar, what you did with some family, as well as some friends, and I have always been encouraging people to do the same.

Harveer Singh (10:26):
Even during COVID times, we have been struggling. A lot of people have been struggling. And I personally think I have been super fortunate. I have grown as an individual, I've grown as a family, I have grown as a professional, but not a lot of people have grown then. So I've always encouraged friends and family to look after friends and family back in somewhere, or some form. So I have some good stories to share as well there that people have listened and have been doing much more. And that's why you see our growth. People are transacting, people are sharing, people are sending money back home because that's where majority of the people are struggling. So it's a great story. Just think about the good that we are bringing around the world, right? It also comes with responsibility, but here are a few different things that I think I want to share.

Harveer Singh (11:14):
And this is public information. If you search for Western Union and money laundering, or Western Union and human trafficking, we have been very involved in trying to prevent. And I will say trying because there will always be things that will fall through the cracks because of the society that we lives in, people are always out to find ways to get around it. But I'm super proud to say that we have been able to stop a lot of those in the past. And we've worked with federal authorities to get some people behind bars, especially when it comes to human trafficking and child trafficking as well. There are some notorious, I would say, corridors which are known for it. And then that's where I would say the AI and the data for good helps us a lot, right? That's why Western Union becomes force for change because we are able to predict, we are able to look at stuff.

Harveer Singh (12:01):
We are able to match the patterns. We are able to monitor some of these activities and kind of take action around it. I have a personal story. I was trying to send money to my wife in Canada in from US because when she got set up here, the only thing you can do is take like $9,000 or less than $10,000 in cash. And I was fine. I was like, "Once you get there, I'll send transaction." And obviously I use Western Union as a preferred way of sending. So it got blocked. It was like, I think $10,000 or something got blocked. And I got a call from the agent and said, "Can you please describe the purpose of this transfer?" And I explained it to him, it's my wife and everything. He is like, "We want to make sure because the first time you're transacting in a larger amount." Typically, I send back like a thousand dollars home, right?

Harveer Singh (12:45):
This was a larger amount. You picked it up a new country, new person, they don't even know what my relationship with the person is. And I was actually happy about it because if it happened to be because there are many type of frauds that if happened to be that I lost that money, I would be very upset. So one phone call that they picked up, that pattern that they picked up, that I've never transacted, I thought that was fantastic, right? I was a little frustrated at first. I'm like, "I'm sending money to my wife." And they're like, "How do we know it's your wife, right?" So unless you share your prosocial profile and all that stuff, and again, the more that we are connected, the more these things will become easier to manage in the future. We are actually very proactive about preventing fraud, but overall, I think it's a very, very super ethical company, we support people, we support our customers and we want to make sure that every dollar that they entrust us gets into the right hands and with the speed that they expect.

Ari Kaplan (13:39):
That's fantastic. And with the human trafficking alone, one of those never ending things, but if you're able to slow things down or even pause for one person for a day, a couple days, let alone 1700 people, that's a huge impact on so many people's lives.

Harveer Singh (13:56):
Absolutely.

Ari Kaplan (13:56):
Thank you for that.

Harveer Singh (13:58):
Absolutely.

Ari Kaplan (13:59):
With AI for good as well, I know having the next generation learn AI is a very important aspect as well.

Harveer Singh (14:08):
We have been using machine learning technology for many years now, right? So the EML, the money laundering laws and the sanctions list, et cetera, they have not new, but what has happened over the last few years that people have become smarter to find ways to evade it. And that's why the technology needs to become smarter. So yes, we can do all the good with AI, but our technology needs to be able to become more and more smarter. One of the things that I think in the last few years we have been able to achieve is the speed at which we transformed our organization to go from on-prem to cloud, which I think in my consulting experience, I've never seen that much of a commitment from our other clients. I don't want to name other clients, but I know right before this, I was with an organization which was thinking for the last three years that whether they want to be on the cloud or not. Again, it's a personal choice.

Harveer Singh (14:58):
I don't want to say right or wrong. But once we decided that the speed at which we executed and we got the leadership support, it was amazing. And that led to us develop more and more, invest more in technology, but not necessarily managing the technology. We want to be able to leverage the best and breed of products that are out there and not worry about, hey, this node went down, somebody needs to go change it. And I need to procure another set of hardware because I need to scale.

Harveer Singh (15:30):
For us, that was a very, very important factor when we decided to do this. And with that, it also gives us additional flexibility to build experiment with new models, without waiting six or 12 months to bring infrastructure. So our use case was simple, right? We want to scale. We want to be agile. We want to be flexible. So that journey was very interesting. And I'm fortunate enough last two years, I've been a part of that journey. I was actually brought in to accelerate the journey. That was one of my first goals is like, "Hey, we are moving to the cloud. What can you do? Sure. Let's do it."

Ari Kaplan (16:04):
Yeah. It's not always easy. A lot of people who haven't gone through that, just say, "Hey, let's move the cloud. Okay. What will we do next week?" Like it's a quick, easy thing. And the end result much more agility. And that's a lot of times what you need. I'm fascinated to hear, what are your thoughts on the benefits of staying agile and maybe a little bit into how cryptocurrency plays into Western Union?

Harveer Singh (16:26):
Yeah. So our positioning has always been proceed with caution. Our stand with cryptocurrency is interesting. Imagine a use case where... And this is primarily hypothesis. It's not like something that I would officially state, but it's a hypothesis, right? If I was to send a hundred dollars to my parents, right? And I started today, which means once it reaches there, it's $96 for example, how will it make me feel? So it's not about cryptocurrency. It's about the use case and the customer sentiments first. I gave Western Union a hundred dollars to transfer a hundred dollars, right? Yes. I maybe I may chose transfer a amount of Bitcoin, but minute it reached, it went down or they forgot to cash it for three days, it went down to like 40, $60 or $70 or whatever, right? I think that it has to be a use case based approach on how we use the blockchain technology to our advantage and not necessarily get tied down, whether it's crypto or not.

Harveer Singh (17:27):
I think it's the technology behind is super powerful. We always maintain that stand, but understands that we are experimenting some of that in the back for the stable currencies as well. We are part of the consortium, which actually talks about stable currencies as well because for us, our commitment is to make sure our customer benefits the most. But if it's at the customer's the receiving end, you can always imagine the havoc it'll create on our business. And no matter what it is, right? It's hard to make someone understand how did it go from like a hundred dollars to $80 when it reached its destination. And also the receiving end, the customer or the receivers, their notion of the crypto could be very different. They might not understand what they just got and how to actually cash it out in some way or some form.

Harveer Singh (18:16):
So there are processes and things that need to be defined. So it's still a long way from becoming, I would say a money transfer per se, but there are benefits to the technology itself to help speed up certain processes in the back, for sure. I, myself was part of a cryptocurrency startup back in 2017, failed one. The reason is Bitcoin crashed during that time, didn't anticipate it. And the startup was very interesting. I was out to create a global loyalty currency where companies can mine and reward their customers with loyalty currency. And that currency can be utilized across the board as a true cross interoperable loyalty currency around the world. So you buy stuff in McDonald's here, but you are retrieving for a Starbucks coffee in another country, why not? Dollar has the same potential, why not loyalty currency has the same potential as well?

Harveer Singh (19:11):
So I think I was a little too early in my thought process, so it didn't really catch fire. At one point, I did actually do it pretty well on the exchange for a few days before it just crashed. But it gave me a lot of experience on what I can and what I cannot do. So I'm going to go back into it at some point. And hopefully with Western Union itself, we can do something in the future with some good regulations. But right now the stance is that we want to make sure that our customers are protected regardless of how the fluctuation is. That's the core of our understanding.

Ari Kaplan (19:49):
Well, before you said, I'm going to go back and do it again. I was going to encourage you. I was going to say, "Harveer, you've got to go do it again, since that's such a good idea." I live for Marriot rewards, but you can only transfer to like United or a handful. And it's like, I want to use it at Starbucks. You can leverage it and then everyone benefits.

Harveer Singh (20:09):
You can help me do it. I'm totally up for it at some point, right?

Ari Kaplan (20:14):
Yeah. I do think you should go back into it since there is a global business around that shared loyalty and incentive. Hoping you could educate us on the Bitcoin, the digital dollar, and then there's a central bank digital currency. Like what are all of those and how do they all fit?

Harveer Singh (20:33):
Yeah. So the way I think about those are Bitcoin is primarily the stock version of your dollar in someway or some form. The stocks go up and down. The value is obviously based on the scarcity of the availability of the actual coin and the technology behind it is actually very fascinating, which is where that it's something that you cannot fix. That is why the role of the middlemen gets eliminated in someway or some form. And that's why it's a threat to a lot of government entities. What is that one thing that tells you that this dollar is valid, right? When you hold a dollar bill? If you look at the writing behind, it says, "We'll guarantee that this will be equal and to whatever the value it says," but if there is no guarantee behind it, then how do you determine that if that's $1, right?

Harveer Singh (21:25):
So I think that's what the technology removes is our ability to rely on a third party to do a mutual transaction. It's almost going back to the middle ages, "Hey, here's a banana, give me an apple." You decide what is the value of the apple and I decide what is the value of banana? One banana is equal to two apples. If we agree that's great, fantastic. And I have five people looking and say, "Yep, that trade happened." But that was simple days. Our financial services have become complex and complex and complex over a period of time. I think blockchain simplifies some of those, but the underlying education is extremely important. Right now, I think people are not educated, they are following other people on Instagram and TiToks and Facebook where people are flashing these expensive cars in yards, but you never really know if they have any of that stuff, right?

Harveer Singh (22:21):
They're making money just posting that. But behind the scene, you really don't know if they actually have any of that stuff. Because it's very easy to figure out who actually has most of it. And there's a very limited number of people. This is the fun part of this technology. It's actually transparent. You can actually go into the chain and see who has the most what's most Bitcoins, right? There are certain big accounts that are sitting there, but haven't been untouched since like 20 years, 10 years or something like that, right? We don't even know who those are. I think people need to be educated, why they're doing it and for what purpose, and be ready to accept the risk in someway or some form. Cash in itself can be valued. Inflation obviously brings the value of the cash down every year or whatever, but it's a predictable value that goes down, but this one is non predictable at this point in time because it's all demand and supply, right?

Harveer Singh (23:13):
How do you feel this person woke up on the wrong side of the bed today, right? I always believe that the people who suffer are the common people in these kind of situations. The institutions are extremely knowledgeable and they can pull out and pull back as people like us who are doing their day jobs, but expect to invest and expect to get a return is when these kind of stupid things happen. The whole thing just comes down in someway or some form. So I would say Bitcoin is like the stock of our currency. The digital dollar in my mind is nothing but a digital form of your dollar bill in someway or some form. And we have been using the form of digital dollar in a while. If you think about it, if you're transferring money outside Western Union, do you actually pay cash or do you pay to your debit card?

Harveer Singh (23:57):
If you don't see the cash in your hand, you are already using some form of a digital currency in someway or some form, it's loaded in your wallet. You think it's there and you have faith that it's there and then you transfer it and you have that faith that it's going to be transferred, et cetera. So I think we are already using it, but the one thing that potentially might be changing is the number of parties involved with that will change and will reduce given the technology is becoming sophisticated, with the blockchain technology being sophisticated. The US government has its own charter of building the central bank digital currency, which means that it potentially will allow the banks to do currency trades as well as settlements much faster without the need for an additional third party in some way or some form, which is where I think the power of the blockchain comes in very, very handy.

Harveer Singh (24:46):
China actually started a digital Yen on already where they're encouraging people to sign up on their app and start using it in someway or some form. And I think as a travel currency, the stable digital dollar could be a very interesting proposition because now I don't need currency exchanges. I can just go and travel around and spend my any as I would. But it's still my digital dollar that I'm spending. Yes, things could be valued differently, but I don't need to see it because I always get confused. Let's say I'm in like Turkey or I'm in Mexico. I have to do this calculation. If that thing just takes out of equation, I think it's a very interesting reposition with how the central currency could work across the world.

Ari Kaplan (25:27):
It'll be very interesting to see five, 10 years from now, how that will all pan out. Do you have any thoughts or insights?

Harveer Singh (25:35):
Well, we already have global currencies, right? So oil, metals, they are some way or some form what you just said. We just don't call them dollar. But today, the oil trades against a dollar pretty much most of the places. So these concepts have been there, but I think it's going to be more centralized in some way or some form going forward.

Ari Kaplan (26:04):
The world is changing. Geopolitical changes, the European Union, the Euro, not the Euro. And then it's also fascinating how Western Union encompasses pretty much every geographical region of the world, Western Union candles different currencies around the world, different financial fluctuations, different regulations. When you go to the cloud or go on-prem, there's all sorts of compliancy.

Harveer Singh (26:28):
You bring up a very good point. I want to tell you the extent in which we have to do it. And that's one of the reasons we have to also choose multiple cloud partners because certain countries would only allow a certain particular cloud partner and they don't want their data to leave that particular country. So again, the Russia, we don't have a cloud partner in Russia. We still maintain an on-prem environment in Russia because the first ping to any services that we provide in Russia has to be in Russia and we cannot take the data out. So we are maintaining a localized infrastructure in Russia, whereas in Malaysia and India, there are obviously AWS and GCP both have certain presence, so we can still be on cloud, but localized the data in those geographies. So India is now becoming one of the countries, which wants the data to be processed in India as well.

Harveer Singh (27:20):
So similar to what Russia wants, but they're okay as long as you are on AWS, but their data gets processed in India. Companies got to find where they were processing customer data in the US, in Europe, which also forced us to look at some of these things on our site. Luckily, we are super compliant with these because of the history of working with all these countries. But these regulations keep challenging us on a day-to-day basis to start to think about an architecture and start to think about how we build a solution, which is scalable, but it's also detachable tomorrow. If I have to really detach something, I should be able to. I cannot not have a detachable solution that I'm creating. So that's why it becomes super complex, but it's also gives me a really good opportunity to see something that I don't think a lot of people get to see the scale at which we get to see things, I'm super blessed to see it as well.

Ari Kaplan (28:12):
Yeah. I love the scale of what you've been doing and you must have a wealth of data assets. There's a monetary aspect, but then there's also the human aspect. And I'm a member of the My WU loyalty program. And I've seen that you've published a couple papers building customer loyalty online, never stops. And another one called building your consumer ecosystem through digital third party access. But you'd love to hear more about the loyalty program or how AI helps you better connect with your customers.

Harveer Singh (28:44):
Yeah. So our loyalty program is taking a turn over the last, maybe I would say eight months. We had an in-house loyalty program that we had built primarily to support retail customers mostly, then it was adapted for digital. But when I came to Western Union, one of the first things that I did was participate and bring in a software service vendor. And the reason for picking or bringing a vendor was that I don't want to build a loyalty product, or a software or platform. I want to build a loyalty product, right? I want to focus on the uniqueness of our product. I want the features that we provide to our customers and then the value proposition that we have our customers. So right now we are undergoing a massive strategy exercise to see how we can make this better, right?

Harveer Singh (29:35):
Yes, you get rewarded for the money that you send or you get rewarded for the fee that you pay. But what about how many times you pay? What about other activities that you do with us? What about the data that you share with us? The ability for us to look into some of your activities that you do outside. Why don't we reward our customers from that perspective to make them part of our daily lives? Right? So today, for example, I'll take you as example. You probably use Western Union once every two months or three months, when you want to send money, right? I want to change that habit. And I want you to come to my app every day because I'm going to have an offer for you in someway or some form.

Harveer Singh (30:12):
With the strength of the customer base that we have, we want to build partners who can offer our customers better value and better products. Why can't I offer you some products that you use on a daily basis? So for example, that's why we launched WU shop today. If you shop through our shopping platform, you can earn anywhere between 5% and 10% cash back that you can redeem for my transfer in the future. You can buy a gift card through us in the future as well. So I want to be part of your daily lives in someway or some form. If I can start rewarding you on the activities that you do, and you start to see your wallet balance going up by then, by the end of the month, you might actually have accumulated enough money to pay up someone in one of those countries. And my mom and she was in India, she was still paying her caretaker at home.

Harveer Singh (31:03):
And it was maybe like 80 bucks, hundred bucks a month or something like that. And I was like, "Mom, whatever shopping you do, just do through WU shop. At the end of the day, you'll have a hundred bucks to take back with you to pay your caretaker." She's like, "Well, that's a good one. Why don't I shop here and then this that gets taken care in India in someway or some form?" That's the kind of one proposition that we are trying to do. And what we do with AI there is we want to be able to understand customer behavior. We want to understand what the customers are doing online, where they're sending money to, how much they're sending, how many times they're sending. And if there are certain occasion they're sending, why not send something else with it, right? So it's all about, how do you maximize the wallet share, but also make sure the customer gets the most benefit out of it?

Harveer Singh (31:50):
We try to reward our customers with majority of the money that we receive from the affiliates, but we are not in the business of shopping, but what we want to do is that creates a stickiness. If you want to go to Walmart before now, you go to Walmart, you would check, "Hey, WU shop, what are you offering me? Oh, nice. 3.2% discount. Excellent. I'm just going to order it from WU shop for example." So, and this has to happen. You have to keep innovating. You have to keep adding. I cannot share yet what we are thinking, but give it another six months, you will see an interesting value preposition, an interesting area of partners that are being onboarded today with this, which will bring immense value. So we are looking at partners from healthcare. We are looking at partners in the real estate world. We are looking for partners in the insurance world. We are also looking at partners which want to do some exclusive things with us because of the sheer amount of customers that we have.

Ari Kaplan (32:41):
So everyone listening shop through WU. I didn't realize that five, 10% in certain situations. And I see AI in other industries like retail and CPG having like next best offer. You bought this, you should also buy this. So I see that as like more targeted loyalty where, "Hey, you sent this to your family back in the Philippines, why don't you also send them X, Y, Z?" That's kind of individual hyper targeted to the people.

Harveer Singh (33:16):
It's also that behavior, right? We want to reward the people for that behavior. So the minute they do that, that is where you know that you established that pattern with them. You establish that this person is more susceptible to do certain things. And when you suggest something, so we are working on that. So you'll hopefully as a My WU member, you'll be kept in the loop over and over again. You'll get some interesting emails around, "Hey, try this out and try that out." So we are doing some pilots with select use of customers. I'll make sure you're part of this, you're a supporter. So I'll make sure you're part of this testing program for this.

Ari Kaplan (33:51):
I'd be honored. How do you even approach tech transformation, staying innovative and the concept of 170 year old company innovating? And that's one of the things you are brought in to do. And even if like how machine learning and artificial intelligence might even accelerate some of that effort.

Harveer Singh (34:12):
Yeah. So some of the things that you have to think about or see when you deal with a company that is that old is, what is the culture of the organization? If you approach it that I'm going to change it, then that is a disaster to start with, right? But you can change it, but over a period of time. If you approach it from that particular angle, I'm going to embrace it, but then I'm going to bring about a change, right? That's always been my strategy is let's embrace and accept. They are used to doing certain things. I am just coming in new. I need to learn the ways before I can show them there is a better way, right? It's not that this company wasn't running before I did join, right? They were doing fine, right? They wanted some new thought process, they wanted some new ideas and that's where I came in.

Harveer Singh (35:03):
But I did not come in to change their business model, to start with. I wanted to make these processes agile. I wanted to make these tech more, I would say, friendly to the business rather than being a black box where you have no idea what's going on behind the scene, right? So build that transparency, design privacy, write upfront, be very open about how we are architecting, why we are architecting certain things, be very modular in nature that, hey tomorrow we want to add on new services, it shouldn't be that I have to open the box. Hey, there's a detachable component that I built, plug that on and we go from there. So I'm actually working on a interesting concept of paper with MuleSoft right now around building a modular financial services organization, where you basically have four quadrants. The first quarter is your core.

Harveer Singh (35:53):
You have to define your core, which is where most of your effort should go, which is most of your innovation go. Money transfer and financial services is our core. So we want to be able to make sure that we are innovating every day, which means that, hey, there might be a better way to do things, right? Money transfer is essentially data transfer behind the scene. You're not really transferring anything. The dollar that you gave at that store, the dollar that I pulled it from your bank account is not the dollar that is traveling, right? How can I be a little bit more smarter around that? So people think that what kind of innovation you can do that? But there's a lot of innovation you can do that. How does settlements process work? How fast it can work? How can we predict and hedge our exposure to different currencies?

Harveer Singh (36:37):
When is the right time to buy certain currency? When is the right time to actually exchange certain currency? When is the right time to make sure that the transfer can happen, et cetera, right? There can be always a fast lane and a slow lane. There can always be a medium lane as well, right? This is where blockchain technology brought this idea around, hey, if I want to send you something, let's take an example of Bitcoin itself. I want to send you Bitcoin. There are three ways to do it. I can pay a lot of fee and instant settle because that is what the miners will get rewarded. Or hey, I don't really care. It'll take two days, that's fine, but I don't want to pay too much. So that particular innovation we are trying to bring inside our world, hey, you want fast settlement? This is what it is.

Harveer Singh (37:17):
If you want you could choose that, hey, I already don't care as long as the money reaches in a couple of days, that's an innovation in somewhere. That's a thought process that is changing, right? Otherwise, it was like, hey, I want to transfer money. Now I want to transfer money. But there are these X number of things that we are designing around it. So that's the core. And then right next to the core is the extended, I would say core, which is now that I have got your attention, what other products can I add on top of which are very equal into core? So in Germany and Romania, this is exactly what we did. We are tagging along an account, which is a bank account where you can hold 16 currencies yourself, right? I'm taking the guess, work out of it.

Harveer Singh (38:02):
You can deposit your salary or you can deposit your checks, you can deposit your tips, you can deposit whatever into your bank account. And you decide what is the right time to exchange that and hold it. So you can exchange it in Romanian RON, and then tomorrow just transfer it whenever as needed, or you can keep it in euros and transfer it as needed. And your loved ones will get it in Euro. They can choose to keep it in Euro or choose to withdraw it in a local currency. Right? So I'm actually giving you choice. And that's what my extended core is. And I'm pretty sure my extended core will become the core very, very soon because this is still a pilot market that, which is, we just finished the pilot, but now we are actually having waitlisted customers onboarding at the moment.

Harveer Singh (38:44):
So hoping that then when we talk again, I will tell you that it's taken off. It's definitely taking off, but I'll tell you that it's now up and running and fully ledged, we have like millions of customers, that is what we would love to see. But so that's our extended core. But then the third quarter in my mind is a white label, which is, hey, I don't want to be in that business, but by customers like it, they want it. I can do a revshare around it. I can make a little bit money. My customers can benefit out of it. And the organization who is servicing us can benefit out of it. And we put a lot of due diligence in finding partners like this. So the WU shop is basically a white label solution for us, for example, right? I'm not in the business of selling any of those products, but I'm in the business of directing my customers so that they can benefit out of it.

Harveer Singh (39:32):
Yes, we benefit out of it. But our customers benefit to most, in 80 or to 90% of the money that we make, we potentially give it out to the customers. We want to make sure they give it out to the customers because that is what is, it's not my core, for example. And then the fourth thing is the referral part of it. So referral part of it is like, hey, you have a great product. I don't want to be in this business, I don't want to have a white, but let's mutually it for customers. Our customers can benefit out of it. I can send you them and you can send me your customers.

Harveer Singh (40:03):
We can be referred to each other rather than wasting money on advertising. And you have an established plan. I have an established plan. We are complimented to each other. Let's refer. So that's my four quarter strategy for this organization in someway or some form. And I think we've started down that path. And hopefully in the next six, eight months, you'll start to see... Next time you'll see an email coming out from us, you'll be like, "Oh, that's what it means. That's the extended core. That's the core. That's what the white label is."

Ari Kaplan (40:31):
Well, I'll be on high alert for emails for now. That's great. And best of luck with that. I think it's definitely the way to go. Our podcast is called More Intelligent Tomorrow, where we talk about what will the world be like five years or even 50 years from now? So just wanted to get your thought. It could be financial related, it could be just general humanity. Where do you see the world?

Harveer Singh (41:03):
I have mixed feelings about it, to be honest. And the reason I have mixed feelings is that I do feel that humanity has definitely taken a step back in someway or some form lately. I have certain examples where I feel that humanity has taken a step back, but then there are certain examples I also have where I think we have exceeded expectation as well. One of the things is that we are now divided into two groups. That's what I view. It's for the government, against the government. It was not very evident. There were gray areas earlier, but right now it seems like it's just completely is like, are you with or are you without? Which I feel is a very detrimental to the society because it encourages way too much positivity or way too much negativity. Life is not about either way.

Harveer Singh (41:52):
Life is about a balance, right? Which is where I think the future is very dependent on the information that is getting fed to us. We are relying on do you really know what's happening in India today? You have no idea. You only look at the news. Did you go there? No. If I tell you a thousand people got killed, you were like, oh, but who's there to check? Yes. There are fact checkers. Again, who's there to check the fact, right? So I think that particular information overload or information manipulation is where I feel that we are under threat in someway or some form. I don't care, which side manipulates, but it gets manipulated both ways, regardless. Both sites for and without are trying to do their best to give us their version of this is what's right for you. As a human being, my intelligence is dying because I'm not thinking I'm just consuming.

Harveer Singh (42:48):
This is where I think the AI needs to be strong enough to actually filter and try to filter as much as possible. Again, I see challenges in that, but I also see that is where the technology needs to evolve to start to filter some of this noise out in someway or some form, like how do you really authenticate something that has been said? How do you really put a seal of approval behind it in some way or some form? That I think will be one of the biggest challenge in the next five years. And you can already see that, right? India is going for elections right now. And I'm telling about India because my state, Punjab where I actually grew up is actually going for elections. There are three parties, like going neck to neck and the statements that are being made in the social media and the stuff that is coming out, there is absolutely way for me telling that's actually happening or not.

Harveer Singh (43:38):
Somebody floated an image of another state and said, this is what's happening in Punjab. And I looked at it clearly, it was a different language. And I'm like a common man would not care, right? Would you care? Yeah, fine, whatever. But that is the incorrect information that is being fed. So, I think the AI for good, which I always keep saying that needs to start thinking in this direction, there's a huge opportunity in this particular field that can start to, I don't call it policing, or I don't call it monitoring because then it becomes a negative. It is more around how does it actually builds a balance in some way or some form. So that's my open again. I know there'll be people who like this and there'll be people who don't like it, but I'm a balanced guy. I don't care what your political affinity is, when you consume information, always think, is this harming you in some way or some form?

Harveer Singh (44:29):
Or is it making you hate someone else? If that is what the intention of that particular information is by definition, it is wrong because that itself cannot be justified in any way or any form. I don't want to bring this topic up, but you've heard this vaccination versus anti-vaxxers and everything else, right? By definition, it is creating hate. It's not about vaccination or non-vaccine, do whatever you feel like, but it actually is encouraging people to hate each other, and that is wrong. How do you prevent that? I think AI has a really, really interesting role to play there. The other thing I feel is that more and more threat to us as human beings will be coming in form of digital threats, which means that one day you wake up and your entire bank account is wiped out in someway or some form. And without a trace, that is why my worst fear actually, to be honest, which is why, if you look at some of these cryptocurrency NFT stolen, et cetera, right?

Harveer Singh (45:26):
You people are putting their millions of dollars in into this, right? How do you really secure yourself at a personal level as well? The banks and the institutions are obviously doing their level best to do whatever. But if you look at all these security companies, so many have popped up because there's so much of that is happening around the world. At one point, I was like, are these companies doing this so that they can actually sell their software? Or there are actually people sitting behind doing some of this stuff, right? I can't make up my mind on that one as well, but that is where the AI needs to be even smarter. And I think that some of that AI needs to come into the hands of the people as well, to make certain decisions and certain recommendations for them. What makes sense for me, how exposed I am and what steps should I take to help myself not waking up in the morning and completely feeling like a loser at some point.

Harveer Singh (46:16):
So I think that those two points are extremely, extremely valid. And the third, in my mind, in the next, I would say 50 years, I definitely see AI playing a really, really big role in autonomous vehicles, autonomous cars. But then there is also the threat of being autonomous wars in someway or some form, which is where if this can happen, that can happen too, right? How do we make sure we stop at a certain level? Human history has always showed us that we do not. Our greed always has exceeded in someway or some form, our eyes are always bigger than our stomachs. We don't always eat for ourselves, we want to eat for everyone. That's the human nature. So at one point that is the threat, that is the balance that need to be found out. I think your topic is absolutely valid. It has to be more intelligent, but stop just there.

Ari Kaplan (47:08):
Well, these are amazing food for thoughts. I fortunately don't have answers to a lot of the challenges. I think these are existential threats for sure. But I do love the two quotes, I'll remember from this forever is more intelligent, not super intelligent or intelligent replacement. And I don't want to just consume, I want to be a thinker or a contributor.

Harveer Singh (47:31):
[inaudible 00:47:31].

Ari Kaplan (47:32):
Those are really deep thoughts. So Harveer, thank you so much. This has been a fascinating conversation.

Harveer Singh (47:38):
Hey, thank you, Ari. Always a pleasure to talk to you. This was a very, very satisfying conversation with you. Obviously, this was a true order for me to participate in this.

Ari Kaplan (47:48):
I appreciate and true honor to have you as well.

AI Voice Over (47:53):
Thank you for joining us on this More Intelligent Tomorrow journey, discover more and join the conversation at moreintelligent.ai. The future is closer than we think.