Untangling bureau data, with Dillon Harindiran

Last week I compared credit bureaus to city walls, defensive structures that, because they’re solid and immovable, can keep the riskiest of customers out, but, also because they’re solid and immovable, can’t always let all the good customers in. Today, we look at those walls from the inside, and it turns out, their design is keeping some defenders out, too.

Ah, let’s say the first step is too high for some to step up onto. Am I stretching this metaphor too far? Possibly. The idea is that lenders and their data are the defenders, and that the process to bring new lenders onto a credit bureau is flawed. Now, if you’re sticking with me, you can think of Dillion Harindiran and his co-founders at Slipstream are building ladders, allowing more defenders to get to the top where they can be useful.

Or, in sensible adult language, Slipstream have built a single, simple API that allows lenders of all sizes - they count start-ups, unicorns, and household names among their early customers - to connect to the three major UK credit bureaus simultaneously, hyperfast and with minimal effort.

You can find Slipstream at: https://oneslipstream.com/

You can find Dillion on LinkedIn at https://www.linkedin.com/in/harindiran/

Or just arrange to speak to him in person via https://calendly.com/dillon . You can also read how he and the team got their cool Calendly links over here https://www.linkedin.com/posts/harindiran_allof-slipstreamsco-founders-have-a-calendlycom-activity-7051568664638021632-51tN?utm_source=share&utm_medium=member_desktop)

You can learn more about myself, Brendan le Grange, on my LinkedIn page (feel free to connect), my action-adventure novels are on Amazon, some versions even for free, and my work with ConfirmU and our gamified psychometric scores is at https://confirmu.com/ and on episode 24 of this very show https://www.howtolendmoneytostrangers.show/episodes/episode-24

If you have any feedback, questions, or if you would like to participate in the show, please feel free to reach out to me via the contact page on this site.

Regards,

Brendan

The full written transcript, with timestamps, is below:

Dillon Harindiran 0:00

A controversial opinion that I've always had, is that I don't think it is possible to build a new credit bureau. I don't think it can be done, because ultimately, it's a distribution game and the established bureaus already have all the distribution. No one wants to join a network unless everyone else is already on it.

And so, the only way there's going to be this real time system where everyone has full coverage, is if the bureaus participate.

If the bureaus don't do this, then this is how it will be and open banking will get better and better and better... and the bureaus will lose market share. And that will happen. So it's really up to the bureaus here to become real time and to become easy to work with.

On a philosophical level, maybe only a philosophical level, you could theoretically get a bureau's onboarding time to a couple of hours, you could make the API that easy to use, fetch and reciprocate - fetching you could do in a couple of hours, reciprocating maybe a little longer - and then you could totally change how the credit industry works.

Brendan Le Grange 0:58

I've got two jobs, if you count this one, and two kids and a wife, so I don't get to watch as much sports as I would like to. But given the chance road cycling is one of my go to's. There's nothing that I like more than, on a July afternoon, to put on a stage of the Tour de France - the mountains, the colours of the peloton, they make for a fantastic backdrop to any day. More than that, though, I love cycling for its tactics and the many ways in which a team could win a race. It's not 100 metres where Usain Bolt wins every time.

The timing comes into it at strategy and teamwork. Do you build your team around one sprinter, give her a full lead hour train and trust her to go mano a mano and went out of the group at the end? Or do you hedge your bets and send somebody in the breakaway?

Almost all of cycling strategy though comes from one simple thing: the slipstream.

Because the rider in front, with her nose to the wind, can be expending as much as 25% more energy than a rider two or three places back, safely ensconced within the peloton. The downside is it makes it harder for an unexpected solo attack.. Or, put into terms that will make sense for today's episode. The lack of a slipstream can make it hard for ambitious competitors to attack the group and shake up the status quo.

Welcome to How to Lend Money to Strangers with Brendan le Grange.

Dillon Harindiran, welcome to the show,

Dillon Harindiran 2:38

Likewise. Well, likewise, it's your show... thanks for having me.

Brendan Le Grange 2:42

I started my first job in credit strategy in January of 2002. Before we talk about Slipstream, and how you're going to be 'fixing credit', let's set ourselves in time and work out in terms of background where you're coming from: so when I was walking into that CapitalOne office for the first time to start my working career, where were you?

Dillon Harindiran 3:03

Well, I would have been six years old, so I was probably figuring out multiplication or something. And my youngest co founder was probably not born yet!

So, you know, various internships and things at various different banks before I ended up at JPMorgan, where I was on the the UK TMT team. So we were doing mergers, acquisitions, IPOs, for all sorts of really cool companies in the UK in the US. That was where my career began.

And, you know, it's interesting, looking back on my interview into JP Morgan, I said, in my first interview, "one day, I'm going to leave and build a business". I didn't know what it was going to be. And I definitely didn't think it was gonna be in the credit space. But I always had that ambition. And so I did my time in banking, I learned a lot there, but building something was was always the plan.

Brendan Le Grange 3:45

When you were at JP Morgan, you were young, just out of university, great job in London, one of the most exciting cities in the world, a young man earning a nice salary, the trap of that can often be that we started spending on those credit cards, we take those overseas holidays, but you very quickly took that leap. So what was it that I guess inspired the move towards Slipstream?

And I suppose you've already mentioned that there was always this ambition to be an entrepreneur, but that gave you that courage to say, I quite like this nice job and salary, but I'm going to try entrepreneurship.

Dillon Harindiran 4:18

I think there's a couple of things I think I knew, I've always been a long term thinker, I knew that if I stayed for too long, I would get hooked and the golden handcuffs would slide on. And before I knew it, I'd be, you know, 65 with a nice big bank account, but maybe hadn't spend my life in the best possible way. So, yeah, I knew I had to leave relatively quickly. But I also knew that banking JP Morgan would expose me to things and teach me a lot in a very short space of time. And that's exactly what it did.

And yeah, I mean, when I left banking, in retrospect, I was slightly naive - I left because I thought I thought I was gonna build a rental business, I was gonna rent Peloton bikes because it was the middle of COVID and everyone wanted a Peloton and naively I thought there was a great business there.

And I was pestering the management team at Peloton - I think I sent somewhere around 15 to 20 emails before the Chief Strategy Officer finally took a meeting with me - and he was like, go away stop emailing. And yeah, the idea was I wanted to rent Pelotons and I realised quickly that wasn't a good idea. And I just quit this very well paid job. But I also knew I didn't want to stay in banking forever. It was always the plan to build something. And on a philosophical level, just the definition of employment in many ways is you're generating more wealth for the shareholders and you are for yourself. JP was a great place to start. I met so many amazing ambitious people there and learned a lot.

Brendan Le Grange 5:43

You left JP Morgan, you started, well, you tried to start this business and realised, okay, this is not going to have the legs I thought it did. Again, very easy to have stepped back into the business world and go back to the salaried life, but you pivoted and you didn't pivot in an obvious direction.

How did the idea of the current form of Slipstream start? How did that move towards credit?

Dillon Harindiran 6:07

In a weird oblique way. We never intended to end up here. We started off, as I said, as a Peloton rental business. It iterated many, many times and eventually became sort of rentals as a service, very similar to buy now pay later. So we integrate into a brand's website and enabled them to rent their products out. Our marquee client, when we when we really hit our peak, was was Bird - you know, the big Silicon Valley inventor of the shared ebike escooter.

We were helping them rent out £2,000 bicycles and we suddenly realised, oh, this is actually a lending business in many ways. You know, it's a consumer credit agreement, you're taking on credit risk, and you're underwriting credit risk. And so that was how we began to discover the credit bureaus. We had no background in credit, no background in lending. And myself and my co founders, were suddenly talking to the big three bureaus figuring out like how you integrate in, what is in a credit file, we were total outsiders.

And we had the, you know, the naive optimism goggles on, about building this rental business. And essentially, the crux of why we shut it down was, you know, a 30kg ebike going out is one thing, bringing it back is actually very complex and painful. And if you can't scale operationally, then you know when you're this BNPL'esque business and you're fitting your unit economics in 3%, 5% of revenue. If you can't get operational scale, you can't get to capital scale, you can't get to capital scale, you don't have a business here.

And so that whole year and a half of building this BNPL business, you know, we had great clients, we were growing, we had revenues, VCs were super interested in us, but we knew fundamentally, it wasn't gonna work.

And the whole time we'd been grappling with this pain point of how do you go live with a credit bureau? How do you ingest bureau data, and the process was, you know, I think it's pretty universally acknowledged that it takes a long time. It's not easy. But you know, my co founders are exceptional engineers, they're used to working with products like Stripe, where you integrated into Stripe in an hour or two, everything is well documented, the API is really clean and easy to use. The bureau's, for many various reasons, are just not like that.

And so we've been grappling with this problem. We understood it really intimately as customers of the bureaus. And we just saw a really big opportunity that isn't being served, where our ideal end customers, the bureau itself. So that's how a bunch of 20 something year olds ended up in the credit bureau space, like we never planned on it but to us, this is a huge opportunity to help fintechs go live use bureau data, but also to help the bureaus to become more competitive relative to things like Open Banking, or to capture this like unserved mass of fintechs who really want to use bureau data, but it takes too long.

Brendan Le Grange 8:47

Yeah. And I think that the value of the data was so big for a time that the fintechs would find a way to deal with it. And it was clunky, but for people that didn't really move the needle for the credit bureau, so I can definitely see how this problem persisted in that world.

Before we get the part of what Slipstream is doing, the nuts and bolts there, we talked about pursuing the dream of entrepreneurship and the pivot - you're now two years into the total journey, about one year into the current format, what was the reality, looking back, of being a founder and that journey versus those glorious expectations?

Dillon Harindiran 9:24

When people ask us, myself, Connor, Rishi, you know, they're like, oh, I have this idea: I want to build a startup. Our advice to people is 'don't bloody do it'. It's so hard.

The naivety is very valuable when you're a founder, like, the fact we'd never done it before and we're slightly clueless at first is definitely how we have persisted. Yeah, it's been a lot harder than I could have ever imagined when I started this. And I think had I known how hard it was going to be, maybe I would have stayed in banking on some level!

I don't know, maybe retrospectively I wouldn't have made the same decision but I personally love it and hate it.

I tell my friends, because they see me in different moods all the time, it's like living on a very precarious you know, there are big highs, big lows, and they happen day after day, right after each other.

And so there have been some real learnings. You know, like the big learning for us is like a deal isn't done until it's actually done. Like how many times have we thought all of us has massive customer, there's massive deal, if we sign this, like, we are set, we're gonna be on a yacht and like, it obviously never never pans out that way. There have been some really big learnings, some big ups, big downs.

But what's what's interesting for us is, we do view this as a journey. And as individuals, as co founders are people who we try to improve, for example, like the speed at which we move, the way we do things is so different today, compared to how it was at the very beginning, building that BNPL business, signing big customers, these were things that, you know, when I first started straight out of banking, relatively young, I was very clueless about and you know, we're a very different team today than we were when we first started. I am a very different person today than I was when we first started.

I wouldn't trade that, but it's definitely been a lot harder than I could have foreseen!

Brendan Le Grange 11:12

Let's move back towards Slipstream, in nuts and bolts terms, what are you looking to do in that space?

Dillon Harindiran 11:19

We want to cut the time it takes for a FinTech or even a, you know, a major bank integrating into a new bureau from many weeks, many months at times - we have people we've spoken to where it's taken up to a year to integrate into a bureau - down to just a couple of days, maybe even faster than that.

And the way you do that is by building the world's best API and integration platform, well-documented, in all the common languages and so on, for any fintech to integrate into any bureau for everything from the sales process, the due diligence, the technical integration, everything.

Brendan Le Grange 11:53

In the old world, they would be driven by the bureau, they would have to say this is what our data structures are, this is what I want from you - two years of historic data - it's got to come through the cleaning process, how much is getting rejected, come back again, okay, we only got six months, and you've got to go to the board as well. What about this? Are we happy to bring them on for less data, and nobody's happy? I mean, this is one of those areas where we're just losing money, it's not competition, it's just

Dillon Harindiran 12:19

It's a two sided problem. It's not just the customer suffering. It's the Bureau's missing revenue.

Brendan Le Grange 12:23

So how does the world of APIs change that interface relationship?

Dillon Harindiran 12:28

Yeah, so the Bureau's have API's. But for example, multiple of the Bureau's use a SOAP XML API. This is an old format, and an old design that no one really works with. And it's very painful to use. And it's very obvious when you talk to software engineers, and my co founders that this is hell to work with. And, for example, if you use our API, you're using this beautifully formatted, well documented JSON API, which is very easy for any engineer to pick up and use.

For example, the response you get when you pull a credit file from one of the bureaus is full of codes and numbers that you cannot intuitively look at and understand what's going on in there. They're tied into formats and designs and architectures that that are not up to par for what fintechs today want to need. Right. And you use our API and everything is already well written out like anyone with no credit background can look at a credit file and just know what's happening in that it's very intuitive. The time spent going to market when you're a FinTech, maybe you've raised a venture round, maybe you haven't months is is literal survival. That is runway and often of startup will have a year to two years of runway to survive. If you're spending six months of that back and forth thing with a bureau of over data, which is critical to what you're doing, you're lending FinTech or whatever, like that is time of survival, that is time going to market getting customers, what we're doing is kind of going well, we're a team that's used to working with beautiful, intuitive, well designed products.

We've also experienced integrating into the bureaus ourselves, so we know how they want things done. And so how do you make the credit bureaus feel like working with stripe? Like how do you make it so easy for a developer, you know, someone who has never worked with a bureau before, like we were novices, to look at it a bureau file and understand it. So intuitively, how can we get the time to market as close to zero as we physically can?

Brendan Le Grange 14:16

There's huge parts of the market that are doing credit in ways that aren't obvious. us a lot of people that just never would have had a route to the credit bureau.

Dillon Harindiran 14:26

Yeah, so we have a spectrum that is unicorns that do hundreds of 1,000s of credit checks a month, all the way down to the earliest of early stage fintechs. That, you know, some of them haven't even launched yet. They know they need bureau data, they need it for their FCA authorization and where the path of least resistance, but the crux of why bureaus don't serve fintechs well, is about the return on time. If I'm a salesperson in a bureau, I have to hit a certain quota each month. And if I sign up a bank, like a big player that gets my quota, you know, I get my commission. If I sign like 10, tiny fintechs some of them might not even make it to the end of next year. Yeah, that's not going to make my numbers and I still have to spend all this time on them.

And so the return on time for the bureau serving fintechs is just too low. And so, at the strategic level, the bureaus understand that this is a bit of a missed opportunity. And they understand that today, what might be the tiny piddly startup in five years time could be the next Monzo or Klarna, or Revolut.

And if you don't serve them today, and they go and sign up with another Bureau, have fun trying to win their business as a primary Bureau five years from now, when everyone else wants their business. And I think what will happen is one of the Bureau's will hopefully partner with us, and then suddenly become very competitive serving fintechs winning a lot of this FinTech demand. And what I think what happened is the other two bureaus will sit up and go, we're losing out. And this is the other bureaus network getting bigger, this is them winning revenue. And if you know, if you think about it, if you use our API, and you win the early stage fintechs, who are very potentially low return on time customers, and you get them on board in a few days, why not do that for a big bank? How does that make you competitive when you pitch HSBC for that primary bureau or whatever? Right.

Brendan Le Grange 16:12

And I think there's also the cultural shift that it used to be credit bureaus were about building big walls, I mean, the whole business model, join the club, big fee involved, big audit programme, six months to a year of sitting down with everyone

Dillon Harindiran 16:25

In the US, they come to your office and audit you in person, like they literally turn up and check that. They're like, can we see the services? Yeah,

Brendan Le Grange 16:33

on site visits? Yeah, it was all about, Are you inside or you're outside. And the people inside the club, were quite strict about who else could come in. These were the big banks that covered 80% of the market 90%, the market said, well, the little guy is not bringing us much extra data. And I think now the way we think about data is changing, it's becoming a lot more open, a lot more sharing.

And that needs to filter through all the embedded processes and things that have been built to survive in a world of big walls, to in our world where data is more open. And of course, there'll be many steps in that direction, but I think you are getting to some fundamental legs in there that say where to how do we bring on somebody with a great idea and 10 customers?

Dillon Harindiran 17:14

You know, it's the gap between what the chief revenue officer wants and what the salesperson on the ground wants. Right. And interestingly, it's also the gap between what the chief product officer wants and the engineers on the front line actually building the stuff. But I think you're right, there are three bureaus each has, well, there are more, but there are really only three. And each has a big overlap on the core big banks, and so on who who use all three.

But wouldn't it be awesome if even the earliest stage FinTech could report to all three?

Because when I talk to FinTech founders, that's actually what they want to do. They see reporting to TransUnion, Equifax, Experian, as doing right by their customers, that helps them build their credit profile that protects them, the vulnerable ones, the ones who are in debt spirals from accidentally accessing too much credit. So I see this really interesting angle of fintechs want to report to everyone because it's right. And then some of the bureaus, the one we're talking to right nowin particular, is their mindset is also completely different.

At the top level, it's not we want to be the siloed. Institution, it's we think this is the future trajectory of the industry, we think that there will one day be a common standard for reporting data two, or three, and that people want to report all three, and we should drive this.

This one particular wants to compete on all its value add not just having the you know, this great network, they want to be all the other things on top of the network. And so our vision is kind of twofold. It's we want to cut the effort and time for fintechs to use and access bureau data. We also want to create this common standard for all bureaus to get data from all fintax. Because it's just good, right? You know, and I think what will happen is, the bureau that drives the common standard, will benefit the most at first, and then the other two will see that they're benefiting and hopefully join into

Brendan Le Grange 18:55

Yeah, and I mean, it's is how you assumed it would work.

I mean, if you didn't have any jading from 20 years in the industry, someone just explained the concept, you would assume you just sent the data, it shouldn't be complicated. If you were explaining to a five year old, they would understand why it was so hard to do all three, when all other data kind of immediately goes up somewhere into the cloud, and everybody knows about it. Why is this data that 40 years now has been the backbone of traditional lending, at least your mass lending in the High Street,

Dillon Harindiran 19:25

The data that fintechs lenders, BNPL, mortgage lenders provide the bureaus with, it's pretty simple stuff like name, address, amount owed, credit limit, and so on, right, like the time, the date, these are like pretty first principles things.

What we're doing is actually really, really complicated, but you can ingest this data and then give it to the Bureau's and you know, the keys format, the Insight format, depending on which Bureau it's obviously a very hard thing to build. But imagine if that real time flow of data via an API became something the bureau started to ingest direct from the API.

Step one is to simplify the integration, simplify the reciprocating of data create a common standard, but eventually I think TransUnion, Equifax, and Experian should become real time networks. You know how many customers have a buy now pay later player goes to all the other BNPL players and borrows money in the same month. And if bureau data is stale by a month, two months refreshed once a month, that doesn't get caught in the system, right? Like you could go to five banks tomorrow, not financial advice. You could go to banks, borrow £10,000 at each bank and just abscond and the system cannot catch that because it's it's it's a slow system. So yeah, like I said, like we to view the bureaus as like the biggest potential customers, but we're solving a problem for the fintechs too.

You know, we see a lot of great innovation in Open Banking. But that's the reason we're not in Open Banking is because we see the bureaus as this like essential thing, making them like Stripe this essential problem to solve.

Brendan Le Grange 20:52

There's certainly a lot of innovation happening here direct it great company that open banking space, rolling out some scores. And I think in that thin file space on the fringes, people are having to use it because they can't get onto the bureau is amazing for that. But there's still going to be for a long time, big old dinosaur of what the last two years of your repayments been.

Dillon Harindiran 21:09

I had an interesting little thought just now.

For us when we think about real time data, and enabling the bureaus to become real time networks, when someone runs a credit check, they pay for that credit check. And it's like a one and done kind of deal. But imagine if instead you went I have the 360 real time timeline of a customer. That's a recurring revenue opportunity, not just the open banking. This is my Monzo card that looks really good. Imagine if you had a low friction 360 real time insight into a customer. I think lenders would pay a lot of money for that.

Brendan Le Grange 21:38

And I see that you also your pricing model is right down to basically individual transactions.

Dillon Harindiran 21:43

Yeah, like we were really trying to solve the problem we experienced when we first started, there was this weird chicken and egg that when you're very young as a FinTech, like very young, you're like Inception Level Young. No one wants to invest in a FinTech that doesn't have FCA authorization doesn't have access to bureau data.

But also, the Bureau's don't want to serve you until you have those things. And the FCA doesn't want to regulate you until you have those things. And you're in this really, really hard place. Okay? We believe in the FinTech model, when a bureau serves a FinTech, they're almost acting like a venture capital fund in that, for VCs, the economics of investing in a startup is that it's a power law distribution, where the 5%, 10%, if you're lucky, 15% that succeed will succeed so spectacularly, they will cover the losses of all the ones that have failed, and some.

And when you're a bureau, it's kind of the same thinking of like, if you don't cast a wide fishing net, you're not going to make the best decisions, like bring down the barriers to entry for the for the fintechs, you know, kill the setup fee, if you need to, like kill the minimum spend, you know, some of them will die. That's that's how it is with startups. But five to seven years from now, you've bagged the next Monzo, the next line or the next revolute. And so pricing is one thing, but obviously, you know, you never want to compete just on price. But instead if you compete on integration, like speed to market and so on that that's a huge role for fintechs. But yeah, like for us, the starting point is we want to help fintechs onboard as quickly as possible, but also as cheaply as possible, because some of these guys are going to be huge. And that will pay itself off for us in the future.

And we're willing to just swallow that cost today, which is expensive, but that's okay.

Brendan Le Grange 23:18

That early market reaction, is there anything that's absolutely happening that you can share in that space?

Dillon Harindiran 23:23

Yeah, there's a lot of interest from fintechs, which is fantastic. It I mean, it validates what we experienced. There's also a lot of interest from the bureaus, you know, I can tell you like you chat to the CPO of one bureau and they're gonna tell you like, everything we're telling, like everything we've just discussed is stuff they know, like they know it's hard for fintechs integrate, they know that they're missing all of this opportunity. And so nothing we've actually discussed is actually that mind blowing to anyone who works at a credit bureau.

And so the bureaus are super interested, the fintechs are interested, and weirdly, the VCs are interested because I think they're hearing from their fintech portfolios... things are going well. But as I've learned, there's always a baseball bat ready to hit you in the face unexpected.

Brendan Le Grange 24:03

So a lot of my audience is in the the credit bureau world. It's been 10 years and myself or in the lending world more broadly. So if somebody listening is wanting to get involved want to learn a bit more about how they could use this themselves, where's a good place for them to go to follow slipstream and to learn more? Yeah,

Dillon Harindiran 24:21

I mean, go on our website. That's www.oneslipstream.com. There is a link to my personal diary on it. If not, my diary is calendly.com/dillon spelt weirdly, di ll o n. And you can stick some time in or email me, the market for credit bureau data is relatively small as in its fintechs of a certain type. And so yeah, if people want to get in touch, email, LinkedIn, whatever. We're pretty responsive.

Brendan Le Grange 24:46

doing so much in such a short time. What is the focus for slipstream, what are the big events we should be keeping an eye on

Dillon Harindiran 24:53

Onboarding customers, as many as we can, as quickly as we can is very much our focus.

You know, we've got our first customers They're all using the API. And obviously, they will be debugging and solving things and building new features. For us. It's it's just onboarding. It's we're not even really focused on revenue. Because like I said, our customers are not so much the fintechs. And, you know, yes, we are serving them. But really, we think our customers are the bureaus. And so for us in the next six months, hopefully a bureau partnership will come out of this. So if there are any people at the bureaus who'd like to get in touch?

Brendan Le Grange 25:27

Yeah, they should be listening. I will check up on all my old Bureau friends make sure they're listening to this one in particular. But Dillon, thank you so much for your time. It's been great chatting, and I think that yeah, you're onto something.

And thank you all for listening. Please do look for and follow the show on your favourite podcast platform and share the updates widely on LinkedIn where lending nerds are found in our largest concentration. Plus, send me a connection request while you're there.

This show is is written and recorded by myself Brendan le Grange in Brighton England and edited by Fina Charleson of FC Productions. Show music is by Iam_Wake, and you can find show notes and written transcripts at www.HowtoLendMoneytoStrangers.show

And I'll see you again next Thursday.