when we are looking at new data sources, we have four acid tests that we run through.

The first is the reliability of the data - so is it from a trustworthy source? Is it compliant? Is it of good quality, all those sorts of factors. The second, as you said, is the predictive nature of the data, is it proving out or identifying the hypotheses in the insights that we expected and is doing something which existing datasets can't? The third one is about its scalability, which doesn't mean it has to be the whole population, but for the actual target audience that we're trying to benefit, does it have enough coverage? And the fourth one is, is it understandable, which I think is increasingly important for us.

And we deliberately set the bar understandable rather than explainable because they're different. My science teacher, much to his frustration, spent hours explaining the basics of physics to me, unfortunately, for me, it wasn't very understandable. But actually, that is a really that's a really important point, because it comes back to this control and transparency. If people do not understand it, then why would I trust it? So for us as organisations with new data and new insights, then we have to be able to help people understand why we're doing it and, and the inferences we're drawing.

So, about the history... actually, Russian retail lending is relatively young, it started less than 20 years ago and it made a huge jump from where we were - just basic stuff 15 years ago - to the state that we have now.

There's a positive outcome of this short history, we don't have much of the legacy technologies, etc, that some of a large international group experience that, take core banking systems, they don't have something that was developed 50 years ago, they're in a more or less modern state.

Yeah, it's definitely shifted away from what can be a very pressurised conversation where you're there, you're speaking to the agent, you're you're under pressure to share those numbers, and maybe sometimes sharing incorrect numbers because of the pressure. Some utility companies, they want evidence of certain parts of the I&E to be shared, like documentation - so our system has that.

So the patterns we're seeing are very interesting, and the journey they go through in order to share that I&E, they revisit their I&E several times before they finally share the I&E with the with the lender, you can see there might be two or three times they're going back and forth to the I&E. They're updating it on our system, doing it in their own time. And then right at the end, maybe after a day, or two or maybe a couple of hours, they then share their I&E so it's a much more fulfilled I&E to then have that conversation with the agents as I said, it's a more mature to conversation. Obviously they share that I&E then with all lenders as part of building that I&E network, which is what we're here to do.

Another indicator that is significant is that BNPL meant around 3% to 4% of the total e-commerce sales in Spain, while in Sweden it is around 20%. So we see significant growth expectations for the years to come.

Then also remember that if I'm a merchant and my aim is to maximise my sales, and suddenly I have to talk to five different lenders, or negotiate with them, I may not have the skills nor the time, to do it. Well, we do that for you. So you just have to focus on what you best know, which is selling through your, your webpage, and we take care of the rest.

And what we shouldn't forget is that, you know, the banks don't want fraud to occur. They don't go looking for that. And they're spending huge amounts of money to defend and protect against it. And trying to balance that with making sure that the service they provide to customers is convenient and easy because you just talked about the frustration of being stopped from getting access to something when that happens, they don't want to inconvenience you, or they also don't want you to be a victim of fraud. We shouldn't forget. And it sounds like I'm a massive supporter of banks, and I am to some degree, but I often read articles and I think the article is completely ignoring the fact behind all of this there's a bad guy, there's a fraudster who's gone out and committed fraud, committed deception. They've taken money out of that, or whatever it might be, the bank didn't ask for that don't want that. I'm just trying to protect against it.

The other thing that we noticed was consumer behaviour changed, which was really interesting. Some of it was taking the subsidies and doing this, some it was taking the payment holiday and doing this, and some of it was just people doing it on their own: but we've seen the largest stockpile of cash going into bank accounts that we've ever seen, of just new deposits that came in over COVID. And so there's this big pot of cash that consumers are holding onto right now.

And I think the provision for credit losses that most of the banks had forecasts through COVID were relatively high and nothing came to fruition, right. They didn't see the link. And in fact, delinquency rates have been dropping, even when deferrals ran off as the freezes expired. We thought, well, you know, the people taking deferrals are probably the people that need it. Once they don't have that option anymore. We're gonna see delinquency rates amongst that population increase... and it didn't. So it did, it's increased slightly, but not to the levels that we had thought. And so I think lenders looked at that brand. And they just said, No, we're ready, we're ready to jump back in and you know, ourselves from a supply perspective, we tighten the reins, you know, it was prudent from a risk perspective what we did, but we're ready to start to get back out and you know, reengage with consumers and lend.

I can answer your question as a consumer. And in general, I can also share my viewpoint on the whole market. No doubt Hong Kong is a very competitive market, the credit card issuers are facing strong competition. For myself, I think I have more than five cards but in general, I will focus on one or two cards that're my major transaction medium. Why I want to just focus on using one or two cards is because of the benefits, because of the discounts, because of the cashback. You can go to different shops or restaurants, you have discounts, you have rebates and then on different purchases, you have also other privileges. That's the major motivation. I think, in general that's true for other consumers in Hong Kong.

And on the other hand, as you're said because they're always new cards coming out with new features, especially the welcome gift, you spend a few $1,000 and then you can get some cash coupon or you can get some premium or whatever - then that will motivate you to to apply for new card, but after that it will just sit there. I think that's a very normal way of life.

Because I've seen a lot of fintechs out there, which had maybe a fantastic proposition, but it lacked a few elements to be really, really successful. We do see some smaller banks that have opened with a very clear digital strategy. They have been very successful to attract customers with a very strong marketing campaign, very visible in the media, and very visible in the streets, etc. What I do think in order for them to be successful, they need to make sure that the bank is not only marketing, colourful websites and apps, but they also need to make sure they integrate the risk policy, the credit risk policy.

I can answer your question as a consumer. And in general, I can also share my viewpoint on the whole market. No doubt Hong Kong is a very competitive market, the credit card issuers are facing strong competition. For myself, I think I have more than five cards but in general, I will focus on one or two cards that're my major transaction medium. Why I want to just focus on using one or two cards is because of the benefits, because of the discounts, because of the cashback. You can go to different shops or restaurants, you have discounts, you have rebates and then on different purchases, you have also other privileges. That's the major motivation. I think, in general that's true for other consumers in Hong Kong.

And on the other hand, as you're said because they're always new cards coming out with new features, especially the welcome gift, you spend a few $1,000 and then you can get some cash coupon or you can get some premium or whatever - then that will motivate you to to apply for new card, but after that it will just sit there. I think that's a very normal way of life.

Well, let me paint a picture of the banking landscape at that time, the banking landscape in the 90s in Ghana, West Africa, we were just beginning to wake up to the advantages that digitization could bring in terms of transforming from manual processes to technology-based processes. And more of automating the manual process, rather than looking at the processes end-to-end and transforming them. Very, very basic accounting, very basic bookkeeping, and all of that.

And nothing really digital as we know it today, because still if you wanted a banking service, you had to go to the bank physically, all that really had to change was that we're able to do a lot of transactions in a shorter period of time. And our books were more accurate.

But by the mid-1990s, I had started getting a little adventurous with what we could do with the technology stack that we had.

… one of the reasons that immigration is so essential to the US economy and to the US labour force is that our domestic population is not replenishing, our birth rate in this country is no longer outpacing the demographic shift as the older generation exits the labour force. And what that means in terms of US population growth is that today, immigration drives over 50% of the US population growth...

There are more people today that move to the US than there are people who turn 18 and enter the financial system… and so not having a dedicated strategy for how to attract and retain the recent immigrant segment is a formula to demographically lose market share over time.

"The risk of giving into temptation is as old as humanity. But there are reasons to think that people today are having to work harder to resist it, particularly when it comes to consumer behaviour. Digital technology has made it easier and faster to buy goods and services in an instant, without the delays of processing that once comprised an inbuilt cooling off period". This might sound like a headline from today's papers, but in fact it was from an article in The Financial Times published seven years ago, almost to the day - at a time when Klarna was around, yes, but only just beginning its global expansion, Affirm was only two years old, and AfterPay only a few months old. Welcome back to How to Lend Money to Strangers, the podcast about consumer lending strategies across the credit lifecycle and around the world.

Yeah, it was about helping a friend to get on the property ladder. She's done very well in her career now, but at that point, she needed some help. And she was like, I really want to stay in my local community, and I really want to buy with my husband here. And she just asked us on WhatsApp if we'd lend, but it was just a lot of money to a group of people at that time. And it was like, how are we going to get paid back because we were all recently graduated. So although that loan didn't happen, that's kind of what inspired the idea for JustLend. I wanted to provide opportunities to people could do what they wish to do.

There's a whole bunch of people out there where the traditional model doesn't work, there simply isn't enough information on these people to make a reasonable credit call...

This is a space where lots of people are working, but very few people can claim results. Because this is also the sort of space where lots of AI propellerheads think they can crack the problem. To some extent, that's true. This is also the classic case where progress is both hindered and aided with experience. It's actually good that some youngster on a beanbag, with long hair, thinks about this stuff completely unhindered by any previous industry knowledge, because that's anyone with too much experience probably thinks too much inside the box. At the same time, with something like credit, you do need to have some other people in there who can say, 'well, yeah, that's cool but you need to take these following five things in'.

That doesn't mean that the thinking needs to be restrained, but someone needs to make it practical in the end. To simply let the same space cadets go mad on this is likely to land you in a heap of problems, if you don't actually understand the lending industry.

…there had to be controls that were appropriate, but manageable in an environment where you've now got people working from home, and speaking to customers on a daily basis. The peripheral technology around the collection systems needed to be able to distribute into those people's homes... What I found really fascinating is, there had already been a shift to digital, and to using digital interaction points – we’ve seen it a lot in the acquisition space, we've seen a lot in the management space, but, historically, the collection space have been very slow to follow up. But what we're really seeing now is a more significant shift to allowing customers to interact through digital portals, and to set up payment plans, and to be able to access information about their accounts so that they can make an informed decision...

we have spent a lot of time recently looking at the infrastructure between data, applying analytics, and we're applying machine learning through our data-driven decision engine, to then differentiate how treatments are applied in the operational systems. And the real key for me is that ecosystem should be fully integrated. And it should be a continuous loop so that you continually are learning from the outcomes that you get from the actions that you apply to those customers, whichever segment they fall into whatever treatment you apply, understanding what's been successful, what hasn't been successful, introducing champion-challengers wherever you can, to test new options, but also to ensure that the quality of the data that you're pulling in to help with those decisions is at the highest standard…

I think the unprecedented level of government support and the way that lenders and consumers alike have been able to help keep the ship afloat, mean that it definitely could have been a lot worse than what we've seen up to date. But to your point, there are still plenty of consumers, hurting out there, and hopefully are getting that assistance as they need it.

So we actually as home prices started coming back up, we saw that consumer started reprioritizing their mortgage ahead of credit card. And what we saw, you know, actually starting back in like the first quarter of 2017, we first see that mortgage overtakes auto as the primary payment. So the phenomenon of mortgage becoming the highest ranked actually started well before this pandemic. And then what we saw in the pandemic, was the separation between auto and mortgage delinquency got even bigger. It's likely due to a number of factors, as we talked about the accommodations, for sure, you know, suppressing that delinquency, but also, you had so many people now working from home that they had to protect their home, they might be willing to maybe let one of their autos go, because they weren't going anywhere, you know, and people weren't taking road trips, and they weren't worried about their car, they had to make sure that they had a place to work that coupled with the home price index in the US.

So that was my focus point is, if you've got nothing, that's where we start… for existing clients, you can just go with the Chief Operating Officer to a branch, have your scoring, talk to the loan officers about the clients, they know them, right, you'd be surprised by how many they have but they know them by name, and test the scoring.

It's not the only that the banks won't give the card to them, because they are too young, they don't have enough income, or things like that, they just don't want to go through the more tedious process to get a credit card… now you don't need to go to any banks, you don't need to produce any documents, and you can get a loan in maybe a couple of seconds.

We are always on the lookout for new data sources, or external providers, and whenever we see something that's interesting, we test it… In the end, it comes down to what can we pay per gini point of lift?

You have to operate in the realms of the known and not in the realms of ‘well, we think this happened’. And that's why you absolutely have to have a champion/ challenger strategy, as difficult as it may be.