Podcast 113: Shivani Siroya of Tala

I am truly fascinated by the work that many companies are doing in the developing world. The challenges are greater but in many ways the rewards can be greater as well. Certainly, when an innovative company creates a successful operation in the developing world the impact on society can be profound.

Our next guest on the Lend Academy Podcast is Shivani Siroya, the CEO and founder of Tala. While Tala is headquartered in southern California its lending operations are focused on the developing world. They launched in Kenya back in 2014 and in three short years they have become the #5 most downloaded app in that country. How they have done this is a fascinating story.

In this podcast you will learn:

  • Shivani’s diverse background and where the idea for Tala originated.
  • What she learned from lending her own money to borrowers in Africa.
  • How Shivani describes what Tala does and why they operate entirely through a mobile app.
  • The countries where Tala is lending money today and why they started in Kenya.
  • Their average loan size, duration and cost.
  • A profile of the typical borrower and how they use the loan proceeds.
  • Why personal data and business data is one and the same in their markets.
  • Their three-pronged approach to underwriting.
  • Why they have created all their technology in house.
  • Their average repayment rate across all their markets.
  • How Shivani views their competition in the local markets.
  • How they were able to become the #5 most downloaded app in Kenya.
  • The kinds of financial products that are on their roadmap.
  • Where they are at today as far as scale.
  • How Tala is getting the word out about their app.
  • Who has provided Tala their equity and debt capital.
  • The biggest misconception that Americans have about the developing world.
  • Tala’s vision for the future.

This episode of the Lend Academy Podcast is sponsored by LendIt Europe 2017: Europe’s largest international lending & fintech event.

Download a PDF of the transcription of Podcast 113 – Shivani Siroya of Tala.

[expand title=”Click to Read Podcast Transcription (Full Text Version) Below”]

PODCAST TRANSCRIPTION SESSION NO. 113-SHIVANI SIROYA

Welcome to the Lend Academy Podcast, Episode No. 113. This is your host, Peter Renton, Co-Founder of LendIt and Founder of Lend Academy.

(music)

Today’s episode is brought to you by LendIt Europe 2017 happening in London on October 9th and 10th. This will be Europe’s largest international lending and fintech event of the year with over a thousand people expected. LendIt Europe 2017 will pack more content and networking opportunities than even before into two full days. This year’s conference features over a hundred and fifty speakers and one of the largest expo halls in Europe with over 70 exhibitor booths. You can register now at lendit.com.

Peter Renton: Today on the show, I am delighted to welcome Shivani Siroya.  She is the CEO and Founder at Tala. Now you might not have heard of Tala, they don’t have a very high profile in this country, but I’ve heard of them and I went and visited Shivani in her office a few weeks ago and I came away very impressed with what they’re doing. So I wanted to get her on the show to talk about what she’s doing in the developing world. She’s basically providing capital in the developing world in ways that really haven’t been done before and they’re truly groundbreaking with what work she’s doing. So we talk about how she’s able to do that, what country she operates in, how she’s able to underwrite these people purely just by using the data on their smartphone, underwrite them all in an unsecured loan in a very, very quick amount of time and with low default rates. It was a fascinating interview, hope you enjoy the show!

Welcome to the podcast, Shivani.

Shivani Siroya: Thank you, I’m really excited to be here.

Peter: Okay, so let’s get started with a bit of background…you have an interesting background, when we chatted a few weeks ago you shared some fascinating things with me so why don’t you tell the listeners a little bit about what your career arc has been like.

Shivani: Sure, so you are definitely right in that I have had a very varied and diverse career up until now. To start out with, I started out I guess in a mix of investment banking starting out very traditionally in UBS doing equity research and became very interested in micro-finance and ended up working in micro-finance for a period of time. Through that process, I realized that I did see some shortcomings within the micro-credit system in the product development side and felt like it wasn’t really moving the needle in changing the financial system. So I did see that it was in some ways improving the quality of life of individuals, but on a very day-to-day basis as opposed to really figuring out a way that we could change the system to include them in the formal financial marketplace.

So from there, I moved on and ended up going and studying econometrics and went to go work at the UN Population Fund and this is probably where the idea for Tala really originated. So while I was at the UNFPA, I ended up working across West Africa and Sub-Saharan Africa and having the opportunity to interview thousands of borrowers in these markets about their daily lives and how they were actually using credit and what I found through those interviews and really understanding that daily life was two fundamental problems.

The first of it was the fact that there really wasn’t a lot of cached data on how customers are using the credit products and how that is contributing to the overall improvement in that quality of life, but because of that lack of data it was also causing a problem that we couldn’t understand at a person level what this person’s capacity or credit worthiness was and which really the way we put that is a lack of credit score.

The second piece is the fact that because we didn’t have that data, we weren’t actually able to customize any of the financial products to meet their needs. So I really started thinking about ways that we could actually bring this kind of daily life data into real life in a more scalable way than doing individual interviews so part of that was as I was in those markets I actually started lending my own capital to these customers and started realizing that the way that I was underwriting was actually based on that daily life and not just purely transactional data and started to realize that a lot of this data was actually sitting on our devices, in our smartphones. So if I could somehow extract that data in a seamless way, I could use that for underwriting and my lending decision. So that’s really how the…kind of what my background was and what led me to really understand the problems first hand and then start to think about what the possible solutions could be.

Peter: Fascinating, that’s really fascinating, Shivani. So let’s just take a step back actually and just get started with what does Tala do today? I mean, you obviously talked about your background and you obviously talked about the smartphone, but when you’re describing what Tala does, what do you say?

Shivani: Sure, so the way that I would describe Tala and what we do is that we’ve developed a smartphone application that allows us to assess a person’s credit worthiness using just the data from their smartphone and we’re able to do that instantly. We then also act as the lender and provide that capital to the customer based on our assessment and we do all of our lending and our servicing ourselves digitally through our actual smartphone application.

Peter: Okay, and you’re not doing this in the United States. What countries do you have operations in today?

Shivani: So we’re currently working in East Africa, we’re in Kenya and Tanzania and then we’re also in Southeast Asia in the Philippines.

Peter: Okay, so why those countries, is there something…obviously it looks like you did some work in Africa, why did you actually start with those three?

Shivani: So our first market was Kenya and when we looked across the globe and started to understand…you know, what were the things that we would need for a product that is based on smartphone data. So looking at prevalence of smartphones in these markets, specifically Android, looking at how would we actually transact with these customers so would it be through mobile money, would it be through traditional bank rails, how would we actually disperse the capital. And we saw that Kenya had M-Pesa which is a mobile money platform and as we know, Kenya is really the leader in mobile money where you have about, I would say, 27 to 28 million people currently have an M-Pesa wallet out of a population of 44 million people.

It’s been about a decade of M-Pesa being in that market. You can just see that the number of transactions and products that are running on that platform are just so sophisticated and really ahead of what we’re seeing in other markets so that was another thing for us. And then I would say, lastly, it’s really looking at the regulatory market…would we need a lending license, what are the regulations there, could we move very quickly into that market and then really thinking about customer behavior so capacity and demand for credit. And so as we did that assessment, really Kenya came out to be the first market that made the most sense for us and we were able to get into that market in about three months from developing the app to actually doing our own lending.

Peter: Wow, that’s pretty fast. So you have an office then in Kenya?

Shivani: We do, we have an office in Nairobi and then also an office in Manila.

Peter: Okay, so the people who work there, what do they do in that office?

Shivani: In our Nairobi office, it is our hub for East Africa so we are servicing both Kenya and Tanzanian customers out of Nairobi and so it’s marketing, customer service, collection, operations and then also some engineering and also QA work.

Peter: Okay.

Shivani: And we do that also in Manila as well.

Peter: Right, right, and when did you launch in Kenya?

Shivani: We launched in March of 2014.

Peter: Okay, so you’ve had a decent amount of track record there. These are loans that people are applying for through the app, can you tell us a little bit about the loans themselves. What is the loan size, what’s the duration, what’s the cost, that sort of thing?

Shivani: Sure, so it does depend on country, but I would say across the portfolio the average is going to be about $50 and then the average duration is about 30 days and the average cost is going to be between 11 and 15%.

Peter: Okay, so you said 30 days so it’s a 30-day loan, $50 and so that’s not an APR obviously, you’re telling me, that’s the cost of the loan itself.

Shivani: It is and so we really think of the product as a fee-based product as opposed to an APR product since the duration is so short term.

Peter: Yep, no, that makes sense, that makes sense. So $50…how much are these people earning, are they…from a US perspective that sounds like a really, really small loan. What are they using their funds for, can you tell us a little bit about the typical borrower?

Shivani: Of course, so the typical borrower, about 60 to 70% of individuals in the portfolio are using the capital for small business purposes and they’re really using the product almost as a revolving line of credit or a digital credit card for the business. So this can be used for buying inventory, buying supplies for a restaurant, it could be used for travel purposes, opening up a new location. We think that $50 is a very small amount, but when you consider, you know, how fast the repeat rate is and how fast the capital is turning over, you actually start to realize what $50 can actually provide to a business in terms of leverage and remembering that it’s 50 US dollars, but in those markets $50 is going much further.

Peter: Right.

Shivani: So we always have to look at it as, you know, the PPI.

Peter: Right, right. So if you were underwriting these people and they’re asking for a $50 loan, what are looking for? You said they’re doing it all on their smartphone so they.. .I know M-Pesa is ubiquitous in Kenya. I read somewhere that it was a very large percentage of GDP runs through M-Pesa in Kenya, but is that the data you’re using? Are you using business data or are you using personal data to underwrite?

Shivani: That’s a great question. It is both, it’s one and the same. So when you’re looking at small business owners in emerging markets, for the most part these are sole proprietors and so, you know, the business data is both business and personal. They’re not actually, I would say, differentiating between those things so when we’re pulling in the transaction data from a person’s smartphone, we are actually seeing how much is going to inventory, how much is going to electricity, to water, to rent, to travel and so we then understand what a person’s capacity is. I would say, the way that we’re underwriting is actually based on three specific areas which are not very different than how traditional underwriting works.

We are first assessing a person’s…you know, whether or not they are who they say they are so can we actually verify their identity, are they fraudulent or not. And so we’ve built a global fraud model across multiple countries that allows us to get down to the individual level and device level data to say, yes, we can verify this person. Once we do that, then we’re actually saying, okay, great, now it’s actually a matter of thinking through what is the credit product and the capacity that we’re willing to give to this individual or the loan limit, I should say.

From there, we’re assessing their debt to income ratio and that is looking at the things you’re talking about which is their consistency in payments in other areas. We’re trying to understand their average income, the average expenses they may have in other things so what are the other outstanding liabilities. And then we’re also looking at…once we’ve understood, okay, debt to income ratio, we understand the limit, now we want to understand what is their willingness to repay based on their behavior and their character and that gets to more interesting things, I would think, like social network so primary and secondary geospatial data, app usage and kind of contextualizing those things to then understand what does this person’s network look like on a day-to-day basis and what do the other people that they are connected to look like as good or bad borrowers.

Peter: So that’s all available on the phone, I mean, I get it in Kenya and I think Tanzania is, I think, picking up M-Pesa as well, I remember reading something about that. What about the Philippines, I mean, if you’ve got one thing that everybody uses I could see how it wouldn’t be that difficult to extract all the intelligence you just talked about there. Do you have systems…what do you do in the Philippines?

Shivani: So you’re right that the volume of a particular feature may actually or I should say category of data may change as the market changes. So in the Philippines we may actually say, hey, the volume of data on app usage or social network data is actually greater than what we see in Kenya or the number of messages that a person is sending might be greater in the Philippines than Tanzania because people are heavier users of texting and WhatsApp and things like that. So that is also something that every time we go into a new market we have to look for those cultural differences in the way that people are using their phone.

But, at the same time, some of those initial things like fraud, those do actually translate across geographies so some stuff we will have to add in, but there is a, you know, sort of a base feature set that we can actually take to every market and that’s more around identity verification. The capacity and the willingness to repay, that does change and so we are building different models for every country. I would just say that we build them so they’re almost configurable to new markets, but there is kind of a set framework you can always take to every market.

Peter: That makes sense. So are you building these models yourself or do you hire companies like Lenddo who I know do a lot of work in developing markets or are you using other vendors, how are you creating these models?

Shivani: No, we do all of this in-house, I mean, really our core competency as a company is that we are a mobile technology as well data science company so we have our own internal engineering, data science, data analytics and credit team.

Peter: Okay, so when it comes to default rates, I’m interested in how your loans have been performing. You’ve been in Kenya now for several years so and these are short term loans so you’ve really got a lot of data and that’s one of the great things about short term loans is that it doesn’t take long to really build up the model based on your own loan history. So can you tell us a little bit about how the loans have been performing?

Shivani: Our average repayment rate across our markets is 92% and I think what excites us about that is that we do that without meeting our customers in person or picking up the phone to do the assessment. It’s completely objective, unbiased and given the fact that we have such a high repeat rate and we’re doing this in under two minutes in a market like Kenya from assessment to disbursement, we’re very excited to see that that repayment rate has stayed consistent so 92% and it’s continuing to get better.

Peter: That’s impressive. I know some installment lenders in the US that would love to have a 92% repayment rate, particularly for some of their higher risk borrowers. I imagine you’re obviously bringing in more and more data all the time. Is that going down or are you finding it staying pretty constant?

Shivani: I would say again, it depends on the stage of a market so in a market like Kenya, you know, that is where we are actually seeing that we have a high population of repeat borrowers so there is more consistency there. In a new market, of course, as we are actually getting better at assessing risk in that market, you’re going to see obviously lower rates initially and then you’ll see a lot more consistency as that repeat population actually starts to become a larger portion of the portfolio.

Peter: Right, okay. So then what happens…you said you’ve got a lot of repeat customers, but obviously we’ve all heard about the payday lending horror stories here where someone took out a $300 loan and ended up paying back $3,000. What is it like over there and for the repeat customers, I presume these are all people who have good standing. You’re not rolling over these loans if they don’t pay on time.

Shivani: Exactly, we do not allow a customer to take out more than one loan at a time. In addition to that, we do both positive and negative reporting to the credit reference bureau. So our goal here is really to use alternative data and informal data to be able to give access to credit to these individuals and take that first risk, but we, again, based on what I learned in micro-finance, our goal here is really to integrate them into the formal marketplace and give them not only access but actually given them choice so that they can then have the ability to choose to stay on our platform or otherwise, you know, get credit from a traditional institution as well.

Peter: Okay, that is very interesting there. So what are their options, I mean, I don’t really know the Kenyan banking system or the market for personal loans. I mean, these are people, I presume, when they come to your platform they’re not sort of shopping around you and three other lenders, what are their options?

Shivani: There are other institutions in addition to ours, that are startups or traditional local players in the market that are offering products to the customers. I would say that in terms of our pricing and the size of loans that we’re offering and the speed at which we do it, we are more advanced and so customers are coming to us as a result of that, but I wouldn’t say that we’re the only ones in the market by any means.

I would say, again, depending on geography, you’re going to see more or less competition, but again, a market like Kenya where you do have a very sophisticated mobile money system, you do have other players that are riding on top of those rails, but we are, right now at least, the number one lender in the market in our space and one of the top five apps in the entire country.

Peter: Wow, so that is going to lead me on to my next question here talking about the scale that you’re at so when you say one of the top five apps, you mean not just finance apps, you mean including Facebook and WhatsApp and that sort of thing, is that what you mean?

Shivani: Yes, you know, when you’re looking at our Google Play store rankings, you really can see that customers don’t consider us just a lending product. They really think of us as a partner and a friend and I think that really speaks to the fact that we are going beyond just credit in a sense for them, we are becoming something of a brand and a relationship and so that is something that, I think, again, we don’t think of it as just a customer and transaction. We’re not trying to make as much money as we possibly can on loan one, but the goal here is really to create a laddered product and a set of other financial products that they can stay with throughout their financial life.

Peter: Right, so do you have those other products now or is this something that’s on your roadmap. Is it just really the one loan product or what are you offering?

Shivani: So currently, we are focused just on the loan product in our markets, but what we’re doing to develop that roadmap is do a lot of user research and testing to really, again, go back to that beginning point of how we started the company which is from, I would say, the person up as opposed to just the ground up but the person up and really understand what do they need.

We put it into three buckets; what do they need, what do they want and what should they do with capital. Once we understand those things, I think we will start to work on products around improving financial literacy. We’re really focused on customer protection looking at products that help them in times of emergencies, growth of their business so small business loans potentially, insurance, savings. There’s a wide variety of products out there, but I think it all keeps coming back, for me at least, in developing that initial relationship very well with your customers.

Peter: Right, so is education part of your offering right now or is that part of the future?

Shivani: That is part of the future so what we’re doing right now is really focusing on the relationship and then focusing on the research.

Peter: Okay, just going back to scale, if you’re one of the top five apps in Kenya and Kenya’s pretty…it’s a decent size country in Africa, can you tell us like how many loans you’ve done or give us a sense of where you are as far as scale because I imagine one of the great things about doing this on an app and there’s no real human intervention, you can scale more easily than the people who have the offline model with loan officers on the ground.

Shivani: Definitely, so we’ve now done over 2.5 million loans and we’ve dispersed over $110 million in originations.

Peter: Okay, and so how many loans is the average customer taking?

Shivani: I would say the average is about 6 per year.

Peter: Okay, and as they reapply for a loan, do the rates get lower, you know, they obviously become lower risk if they’ve paid you back three times. That’s a much lower risk than someone who is brand new, I imagine.

Shivani: Exactly, so we do have a repeat model for our customers and so we do customize the loan offering as they continue to stay in our system. So, again, it’s a laddered product so it’s not only that the size does change and then our assessment of them continues to improve as they’ve actually been in the system longer.

Peter: Right, right, got it. So then how are people finding out about Tala? I imagine you’ve got pretty strong word of mouth, but how are you marketing yourself in these countries?

Shivani: So it’s very similar to the way we find out about applications. Our customers are seeing digital ads so they’ll see Facebook, they’ll see Twitter ads, they’ll also see ads on their own local networks. In additional to that, like you said, about 50% of our customers are coming in through referrals and word of mouth as well. What’s really cool, I think, on the referral side is you really get to see that strong correlation between repayment rates and other good or bad borrowers in that person’s network.

Peter: Okay, so let’s switch to the other side of this business and that’s talking about the funding of these loans. Obviously, this is not a marketplace when you’re doing things that quickly, you’re a balance sheet lender, I expect, but who is providing the capital to loan this money?

Shivani: So we are again acting as the lender here so it is our balance sheet capital that we are lending out. We’ve raised our own debt capital to fund the portfolio and then obviously raised equity capital for our own operating expenses.

Peter: Right, and can you share any of the names of the companies that you’re working with?

Shivani: Sure, so on the equity side, this information is public, so our seed round investors were Lowercase Capital, Collaborative Fund as well as Google Ventures. Our Series A was led by Data Collective and our Series B, most recently, was led by IVP. We also had Ribbit Capital participate in that as well.

Peter: Okay, that’s some blue ribbon names there. What about on the debt side, on the…you know, you’ve got a warehouse line or some kind of line, I imagine.

Shivani: So what we’ve done there is a mix of high net worth individuals participating in facilities as well as venture debt capital so it’s definitely been a mix of those two things.

Peter: Right, okay. So we’re almost out of time, but I wanted to get a sense of…when you’re talking with other Americans and you’re talking about what you do in these developing countries, I’d be curious to get your perspective on what is the biggest misconception that people have about these sort of underbanked populations that you focus on in the developing world?

Shivani: I think the biggest misconception is the value that these customers represent and the opportunity so I think that there is this misconception that really the only way to serve these customers is in person or taking a grass roots approach or through traditional micro-finance. I think what we’re really showing is that a company that is working in the US, in the same way as other Silicon Valley companies, is going after a massive opportunity, but using data science and the most innovative technology that’s out there.

So we think of it almost as our mission as a company is by any means necessary use whatever data is out there to really prove the capability and the potential of these customers in these markets. So my goal is really to put all of these individuals on an equal playing field with the rest of us and so the same way that we score a customer in Kenya or the Philippines, we can actually use that same technology to do that here in the US.

I think that’s one thing is that, I just want to change that perception of risk and that misconception of this population. I think the other thing though is to remember that each of these markets, they are very different so there’s not this one-size-fits-all solution. We’ve heard about the Uber playbooks and all these things, but it’s also remembering that part of credit and financial services is that relationship you build with the customer and really there is no financial brand out there in emerging markets that has that relationship with these customers. So part of that is developing the trust and then figuring out how to use data to create a very customized solution and develop your product based on that.

Peter: Right, so that brings me to my last question here and that is: where are you taking this, are you focusing on geographic expansion throughout the developing world? As you just said, this could apply to the US, the technology you’re developing really could apply anywhere, but what’s the vision? Is this worldwide domination and you’re going to become the global brand that everybody knows in the developing world, what is your vision?

Shivani: Sure, I really want to take it back to the vision that Tala has. Our vision is really about bringing financial access, choice and control to the underserved globally. So, like you said, yes, that does relate back to the US eventually because there is a huge population here that I would say is underserved and that data is not really being used to create these financial products in a customized way.  So we believe that we want to deliver this value in whatever way does make sense so we are starting with emerging markets because we think the opportunity size is bigger there right now and we can move more quickly into those markets, but eventually, we hope to be able to bring those same learnings and those same solutions back to more developed markets as well.

Peter: Well, Shivani, it’s fascinating what you’ve been able to achieve in a pretty short amount of time and I certainly wish you all the best. Thank you very much for coming on today.

Shivani: Thank you so much for having me, it’s been a really great conversation.

Peter: Okay, see you.

Shivani: Bye.

Peter: I said this before and I’ll say it again right now that for fintech to truly reach its potential, it needs to make a difference in people’s lives in the developing world for the underserved and that is what Tala is doing and I applaud them for their efforts here. Companies shouldn’t do this because it’s the right thing to do which is what I believe; they should do it because this will lead to higher profits.

I truly believe there’s more opportunity in the developing world than there is anywhere else. Most people just ignore it because it’s too hard and frankly, it is obviously a lot more difficult than setting up a company in the developed world, but the potential is greater. There are billions of people who are underserved, there are many people moving from poverty into the middle class and it’s the fintech platforms that are going to help them make that transition and I think the opportunity is enormous.

Anyway on that note, I will sign off. I very much appreciate you listening and I will catch you next time. Bye.

This episode was sponsored by LendIt Europe 2017, Europe’s largest international lending and fintech event. It will be held in London on October 9th and 10th of this year. To find out more and to register, just go to lendit.com.[/expand]

You can subscribe to the Lend Academy Podcast via iTunes or Stitcher. To listen to this podcast episode there is an audio player directly below or you can download the MP3 file here.

Peter Renton is the chairman and co-founder of LendIt Fintech, the world’s first and largest digital media and events company focused on fintech.

LendIt Fintech conducts three conferences a year for the leading fintech markets of the USA, Europe, and Latin America. LendIt also provides cutting-edge content all year long via audio, video, and written channels.

Peter has been writing about fintech since 2010 and he is the author and creator of the Fintech One-on-One Podcast, the first and longest-running fintech interview series.

Peter has been interviewed by the Wall Street Journal, Bloomberg, The New York Times, CNBC, CNN, Fortune, NPR, Fox Business News, the Financial Times, and dozens of other publications.

Subscribe
Notify of
0 Comments
Inline Feedbacks
View all comments