Just to rewind a little bit, this whole BNPL craze was exploding. I hadn't seen anything like it: interest free credit didn't really make sense to me and was very excited about the prospects of this business model. But at the same time, a lot of the people I spoke to in banking and lending were just like, the last thing South African needs is just more unsecured credit being piled on top. This isn't Australia, isn't Sweden where Klarna is from.

So I did take that to heart and looking into the BNPL space realised that in countries like Brazil, Mexico being two significant examples, point of sale interest free instalments have been there for ages offered by most retailers. Yes, it's now become tech enabled. But in those markets, it's done very much off the back of a credit card. And the bank is the key issuer of their credit, they understand the consumer and what they're earning. And the thesis was that the South African use case was far more similar to those markets, developing markets, high interest rates, very disparate levels of income. In some places, you'd argue over indebtedness of the middle class, as opposed to a massive need for financial inclusion, which is the narrative.

Yeah, and I just thought that that product would fit so well here, we just have to find a way to technically adapted and created here.

And then the other piece of the training that I do is I work with senior analysts as they begin to step into their first leadership roles. Because the other thing that is hugely bought by I found very relevant to my world was that when you start leading, it's as if someone goes here, the keys to the car, if you go, you mean you want driving lessons to beat after school drive.

And a lot of us analysts types are super introverted and really fact oriented. And we like our processes. And we don't necessarily do the soft skills particularly well. Yet we all respond really well to those being done well, but they don't necessarily come naturally to us. And I include myself in that. But it's something that we need to learn and be aware of how do we communicate with people up the chain? How do we take analytics and when someone says, but I don't understand. As an analyst, our tendency is to double down on the detail. And when you know what you're looking for, you can literally see people lose the will to live because they don't understand and that isn't helping, and they're not number oriented. So just make it stop.

.I think it was partly due to a chance that we eventually built Taran DM because it was at the beginning of 2020 and we had two interesting projects. So it was like a lot of potential and one of them was a decision management platform, like some customised one with scorecards for a new fintech. And the other one was a platform for a large German automotive company, which was supposed to optimise their supply chain. Middle of February we were flying to the German company, to the exporter. And we had a really good session with the management team, getting a lot of ideas on how to move it further. We were super excited about it, but as I said, like it was February 2020.

So when COVID was spreading within a few weeks, those guys stopped answering our emails and then we got back to them later, they said sorry, but our supply chains have gone haywire because of Covid, we cannot do anything a few months or maybe even a few years!

So that project got killed. And then we ended up with the other second big project, which honestly was, I think, a better fit for us because most of the team was coming from the finance, we had like the experience with infrastructures doing like real time decisions, whether it was credit risk, whether it was the high frequency trading was a large part of the team came from actual credit risk teams that may have been using the tools such as FICO Blaze or Experian Power Curve. We know the the strong points, the weak points, so big up to the I would say drawing board and we thought maybe there is like a market opportunity here or let's you know, let's basically build something.

Whereby from the report we see that 7.5 million people in Tanzania do have bank accounts. And with the presence of mobile money operators, we have 24.4 million mobile money wallets currently opened by these telcos talking into the lending space.

In particular, the banks still dominate the biggest share in terms of value, but the market has seen a new narrative of digital loans, which is mostly dominated by MMOs and fintech players through their micro lending services, which still requires banks collaborations by funding for regulatory approval, as well as managing the provision side with the help of the FinTech players will bring onto the table, the scoring and big data capabilities.

CreditPy is including some functionalities regarding to develop credit risk scorecards, a PD model, basic data analysis, and it checks the informative variables in an automated way to determine which features is going to be passed to the predictive model. It also generates an automated model framework that is actually searching for the best predictive model across the different feature sets that potentially can be used during the model development.

And after this, there are actually many functions that has been defined to create the rating scale. And also, after creating the rating scale, its offers to do some validation, like univariate gini check, information value checks, basic multicollinearity checks, stability checks on the futures to see if there will be any drift on the predictions on the auto sample set bit applies a basic rate of evidence transformation on the data.

And finally, it allows the user to validate the created rating scale, predictive power of the model and the calibration.

But effectively, if you think about it, there is no company in the world, maybe just a few, whose whole businesses is data, noone really just creates data, noone really just handles data. However, every company, whether you're logistics, retailer, bank insurance, your mom and pop shops on a corner, they all deal with data - you've got prices, you've got sales, you've got measurements, if you are building a house.

However, as you said, often in organisations, they try to put a silo around data, they say, I have a Chief Data Office, I have a data function, I have specifically data processing, and data is a bit like the blood of an organisation. It goes everywhere, however, because it goes everywhere, you cannot just silo it somewhere. Of course, you need to have some patterns, some standards around data, but every part of the of a business has to be responsible for the data.

Australia for whatever reason, is unnecessarily divergent and complex, not in the pricing aspect of the lending, that is fairly competitive, but when it comes to the approval part of the process, there is a huge amount of customer confusion as a result of that.

And one, if not the highest penetration of mortgage broker deals as opposed to direct lender deals. The way a traditional broker would solve that level of complexity is through deep knowledge and expertise and experience in having done lots of deals, and usually a close working relationship with a small number of banks. And so the reality is, and industry data supports this, that most brokers will typically use one, two, maybe three lenders for overwhelmingly 80% of their loans.

When we looked at online mortgage broking we looked at how might you make that better and different investing in incredibly high quality tooling for the broker to turn that broker to a superstar?

So if you were to think of some of these African countries and then go into Rwanda, you will understand that the credit growth and maturity is fairly new outside of South Africa, most of these African countries had predominantly been cash-driven markets, it's not to say there wouldn't have been a form of lending or products that would have been there. But if you think of the South African credit market or maturity curve, you know, retail, for instance, people go and buy clothes on credit, you have never seen that in any of the African markets, they get a shock to hear that you buy clothes on credit. So there's still a lot of whitespace, there's still a lot of opportunities for better products to come into play. So personal loans have predominantly been the default lending product that we've seen in the market across all the financial institutions in Rwanda.

But now mortgage, your typical mortgage, right? It's starting to emerge as a product, that historically people would probably just get a personal loan and go buy a plot of land and try to, you know, use their own income to build a house for themselves. What you have started to see off the lead is vehicle financing starting to come up in Rwanda. I'm not sure how close you are to the VW project that kicked off, I think, a couple of years back where they had a factory and wonder and assembling factory and wonder, because outside of South Africa, maybe with the exception of Botswana and Namibia, most of these countries do a lot of great imports. Right? So you find a lot of Japanese cars out here, right? So with that said, vehicle finance, the process of it has been much of a challenge because the bank doesn't know the vehicle, they're financing, they don't have confidence on the quality of the vehicle, right?

So that's why we work with Airtel to say, let's bridge this gap. Let's try to be the boundary between the lenders and the borrowers. And the only way you can do that is having standardised credit score.

So Airtel provides all the transactions when you buy airtime on your phone, when you buy data, pay a bill, you know, that information tells us probably you have some responsibility in your house, you're moving around, probably you have some kind of mobility, are you having so many people different one sending to you, tells you you're making sales.

But if you're just receiving from one person, we can can infer probably your law student getting some update from the parents.

And where I found that problem statement, that focus, was actually when I started working with community finance lenders, specifically, an experienced that really exposed the problem, to me that exists within consumer lending. And that is that very little is given to the customer management side of the credit lifecycle.

And I've had the opportunity to sit within the community finance lenders office, I mean, this was right on the front line. And I remember there was a lady that came in, and she had lots of children with a, she was stressed because she was in arrears. And she come into this lending office to arrange an arrangement with the lender to ensure that she could stay on track with her payments. And I just sat there observing, and she sat there and she was getting more and more stressed, as the advisor was saying, Well, can you afford this much a week? Can you afford this much a week, and having the awareness as to all the data behind the lending decision, and everything that they had about it, I just felt like, there's got to be a better way to engage this customer and use this information to help her make sounder financial choices.

That was my lightbulb moment.

That's why you're hear 'the problem is the deposits', but people don’t need to have to high deposit. Not really, because they only need a high deposit, because you can't lend them enough, because you put all the risks to them. You know, if you change your product into a long dated fixed, you protect the borrower, you can lend a higher amount, they need less of a deposit. Right? It all depends on how you explain things.

And this is what other countries have done for a long, long time: Denmark for over 200 years, the Netherlands for 40/ 50/ 60 years, the US obviously, after the savings and loans crisis, they started protecting borrowers by making long dated fixed rate mortgages the 'cool product', and that protected them over the crisis against these shocks. And that in turn, helps the economy.

I agree, I think is a game changer.

And I think what's interesting as we as we looked at the financial inclusion score more we realised how important lifestyle was as well as behaviour. And that sort of led us down to something were developing to the moment which is really based on geographical havior and customer blueprints for more targeted marketing strategies, we derive a lot of this information directly from the mobile, someone's behaviour on that phone and their choices and their lifestyle patterns and gleaning all that information from the mobile, which is all anonymized data at one point.

It's still dominated by your credit card issuers.

So most people if they have Apple Pay or Google Pay, they have loaded their credit card and that's probably the most popular form of payments, but these QR payment providers who have their own mobile apps is very, very popular. And it's accepted everywhere. And while they still represent a very small share in terms of total purchase volumes, they are by far the fastest growing, and that is why issuers are very, very concerned.

And these QR payment providers are also going into that credit space, where they're offering a small credit of maybe $500 to $1,000. But they're using very basic information - just your name, phone number, email - so it's much, much faster and easier to apply for that new QR payment credit versus a traditional credit card.

First of all, most of the brick and mortar banks just don't understand, at a granular view, which products and which segments are really making them money.

In general, they can tell us, sure, but they have a lot of different customer segments, right, different products and sometimes they struggle to understand where they need to adjust.

So in this sense, our advanced analytics is helping banks to understand how does existing products work, and then give a different insight and basically all of these just allow them to be more agile, and to run business in the more well controlled and data driven way.

"I don't understand why I have to take credit out to get credit. Why don't you just look at my bank statement, you can see how I'm behaving?"

Well, that's great. That's exactly what we do. At Direct ID we focus on categorising transactions that have particular interest risk decision makers.

So what we were able to do is to bridge the gap from unsupervised learning, where we don't have any labels, to supervised learning, where we do have a label.

There is an adversarial goal here: if you increase the limit, you have a potential profit from the person using the credit limit but you have a very real immediate hit to your provisions.

So now we needed some sort of modelling that didn't just give us whether to increase or not, but which would also give us the optimal value of that increase. We don't just give you a limit increase, we give you the one that minimises the value at risk and also the one that has, in terms of expected value, a profitable margin.

And that was a big, big shift, right? Because all alternative lending platforms were profitable, when rates were at zero, they started to go up, most of them are going past. The reason I think that we are noticed. And obviously we didn't do this, anticipating that this was going to happen. But rather, it was just our strategic focus at the time, we decided to cover the full spectrum of working capital of our clients, right?

So we started as, hey, we get cheap money, we lend it digitally to clients. But then as we interacted with those clients, and we recognise that they had a broad range of needs, we decided to cover the whole spectrum. So we moved from, hey, how can I give you money, to how can I simplify your operations through money, right? And how can I make your financial operation simpler, smoother and more effective and efficient through money

So what is driving the default of an SME on a loan? We focused on three potential drivers: the firm's resources, board level factors, and the loan attributes.

I don't want to bore people going through all the variables that we considered but we have three categories of variables and within the categories we have a bunch of other variables. So what we found basically is the following an increased amount of financial resources and increased size of the board and a longer board's tenure, all these three elements are decreasing the default rate of firms. This is quite reasonable, right?

The size of the board is maybe a bit less intuitive. We are talking about the board size, we're not talking about the firm's size, which plays an extremely important role. But our focus was on the board size wich speaks to corporate governance being important there is a debate about corporate governance, many people make it more complex, but that is helping sharing the responsibility. So you may come a sounder, maybe a more bulletproof decision.

It's another excellent question, because I think there's a lot of discussion about big data and AI and machine learning. And they go together well, big data and machine learning, but they're not the same thing.

A lot of what gets done with machine learning in our industry is applying very nonlinear methods to the same old data. In fact, everything I've talked about so far has been more intelligent use of the data you've always had, Building Better models of your business and of your product.

If you have unique data, that's great. And often we find unique datasets in finance companies, where they're doing some kind of specialty lending. You know, one of my favourites for a long time was a group that was looking at point of sale loans for cruise ship tickets.