Our news and views relating to Data Analytics, Big Data, Machine Learning, and the world of Credit.
At Principa, we engage with clients and organisations across the entire credit lifecycle and track the focus of the South African credit industry. For nearly ten years the focus has consistently been in the collection space, but recently (since early 2021) this has changed and a large number of our clients are focused on acquisitions and originations.
Although not a new concept, very few lending organisations have deployed a true multi-bureau strategy (MBS). It is however talked about fairly regularly, but often dismissed as “too hard” or “not important enough”. So why should you consider a multi-bureau strategy? What are the key considerations? How do you go about deploying a MBS? This blog hopes to address all these questions.
A scorecard is a mathematical model that is used to predict a certain outcome. In credit this might be the probability of default. The information used in a scorecard can vary, but common fields include demographic characteristics (e.g. age of applicant, number of dependants, time spent in current job) and credit bureau data (e.g. number of personal loans registered to applicant, worst arrears status on all accounts in the last 6 months).
With the large drive in account origination towards digitisation and automation, in our experience much focus has been on bedding down omni-channel capability. But the real unsung hero of an originations’ project is the API hub. In this blog I unpack API hubs and APIs available in originations.
DecisionSmart decision engine drives innovative App that assists Saudi nationals with home ownership.
During the COVID-19 crisis, the media has focused much on the weak economy and stressed South African consumers. Figures show an increase in unemployment and for those lucky to be employed, many suffered decreased earnings through salary cuts. All this points to a highly strained economic environment.
The South African credit bureau TransUnion recently released data on the performance of various different products within the bureau in their ”Quarterly Overview of Consumer Credit Trends” for the third quarter of 2020. With the COVID-19 crisis, 2020 was characterised by a severe reduction in account originations and payment holidays in Q2 with a high increase in non-performing accounts in Q3 as payment holidays ceased and stressed consumers failed to pay their accounts.
It is common knowledge in the industry that the credit risk assessment of a consumer applying for credit is far less complex than that of a business that is applying for credit. Why is this the case? Simply put, consumers are usually very similar in their requirements and risks (homogenous) whilst businesses have far more varying risk elements (heterogenous).
Principa Decisions (Pty) L Globally there is immense pressure on collections and 2021 is due to be another year where collection departments will be forced to improve what they do and how they do it. A variety of themes have emanated over the last few months, and collection departments are focusing on how to better engage the debtors as well as the introduction of sophistication through themes such as digital collections and advanced scoring. Scoring in collections webinar In November 2020 Principa ran a “scoring in collections” webinar in partnership with CIS Kenya (equivalent of South Africa’s SACCRA). During this 1 hour and 20-minute webinar we covered a wide range of themes surrounding scoring in collections. This included how to introduce scoring into collections, the different types of scoring and then some of the bigger themes such as how machine learning might help your collections. If you’d like to know the answer to the last point, please download our free whitepaper which details how Principa have successfully deployed machine learning models and a machine learning platform at the largest debt collection agency in Africa. The list of topics covered in the webinar are the following: Collections 2020 Different types of scorecards Using a collection scorecard within a collection strategy Bringing all the models together Machine learning Contactability model considerations Settlement model considerations Q&A To watch the full webinar click here. If you’d like to discuss your collection or scoring concerns, please be in touch with us below.
Principa Decisions (Pty) Ltd.
Principa’s FinSmart has become the industry’s go-to solution set for end-to-end credit management. Our credit risk management software products reduce risk and improve profitability by streamlining processes, increasing efficiency, and automating data-driven decision making across the credit lifecycle.
Principa’s onboarding chatbot solution; Atura allows lenders to engage a customer effectively through an application process while accessing necessary data and decisioning calls using Principa’s SmartSuite software. The digital revolution “Digital” has been a financial services buzz-word for some time. Most South African lenders Principa works with have been working hard to adapt to a digital existence for several years. Some have been successful, others are still working on the challenge - and most have only partially adapted.
The time is NOW for model validation and adjustment. One of the major premises used in credit scoring is that “the future is like the past”. It’s usually a rational assumption and gives us a reasonable platform on which to build scorecards whether they be application scorecards, behavioural scores, collection scores or financial models. That is reasonable until something unprecedented comes along. You can read about this black swan event in our previous two blogs here and here.
Payment holidays have been used throughout South Africa and around the world to help alleviate the economic stress during the COVID-19 lockdown. In this blog we look at some of the steps taken internationally and by some of South Africa’s major lenders (specifically in the consumer space).
If 2020 was not hit by the COVID-19 global pandemic, many were touting 2020 as the year of alternative data. In the credit assessment world, data has typically incorporated demographic data and credit bureau data (where available), but now we are seeing alternative data playing more of a role namely in cellular behavioural data and psychometrics.
Principa employs a variety of best-practice credit scorecard building techniques including mathematical programming, regression modelling, optimal segmentation-seek genetic algorithms and reject inference parceling, amongst others. Through our credit risk scorecards businesses can look to improving their credit risk decisioning by 5-30%.
In a previous blog, we looked at assessing your credit models and the challenge of building and deploying models representative of the COVID-19 crisis. At the crux of the challenge was the fact that:
“Unprecedented” is a term with which we’ve all become quite familiar over the last few months. COVID-19 has changed our society and our economy quite drastically. In predictive analytics “unprecedented” has far reaching implications – simply put it’s difficult to build models when we do not have data that reflects similar trends to what we will expect moving forward.
Effective account management strategies can prove highly profitable, particularly in a competitive marketplace where acquiring good customers is a challenge.
In our two previous blogs here and here we looked at how an effective account management strategy can result in profitable decisioning. In this blog we look at what is required to deploy account management strategies.
In our previous blog we introduced the value of effective account management strategies. In this blog we continue to run through different account management decision areas. In our next blog we will discuss what is needed for an effective account management strategy.
Onboarding good new customers in a very competitive market is difficult and not always that profitable. That is why prudent credit managers often look to account management strategies to reap the greater rewards. Once a customer is on-boarded and has proved herself to be an exemplary re-payer, offering her money is more of a sure-thing for repayment.Behavioural scorecards offer far better Ginis than application scorecard. This means you can afford to be more aggressive in account management.
In our previous blog, we covered five steps to help lenders avoid application fraud. Application fraud has been an increasing blight on credit books particularly with the growth in digital channels. In this blog we explore another five fraud-mitigation steps.
New technology gives the promise of greater enablement. But some of the shrewdest entrepreneurs understand that opportunity comes from the unintended consequences of new technology. So, let us take digitalisation of the loan application process: the opening of digital channels has enabled lenders to service their customers 24/7 and through APIs integrate with a host of sophisticated services. However, the advent of the digital channel has meant more opportunity for fraud. The question to ask is:
One of the major premises used in credit scoring is that “the future is like the past”. It’s usually a rational assumption and gives us a reasonable platform on which to build scorecards whether they be application scorecards, behavioural scores, collection scores or financial models. That is reasonable until something unprecedented comes along. You can read about this black swan event in our previous two blogs here and here
This is the second of a 2-part blog. You can read the first blog here.
One of the basic principles of credit scoring and modelling is that the “future is like the past”. Whilst robust credit models may be calibrated on multiple time periods, this assumes that trends in the past represent what is going on today. COVID-19 is a black swan event – meaning in the modern day it really is unprecedented. If you have never come across the term black swan, or if you have but no idea the origin, I recommend taking two minutes to read its really interesting etymology.
With a recent judgment being upheld in favour of the National Credit Regulator (NCR) against Shoprite Investments Limited, we thought it would be a good time to re-look at the process of affordability assessment.
At Principa, we are passionate about new ideas and product development. 20% of our revenue is ploughed back into innovation. One of our key areas of focus is machine learning where we have built machine learning solutions in both collections and the customer acquisition space. While our focus is primarily on credit risk and customer engagement, we are always interested in how machine learning has gained traction in other industries.
Whether you’ve been involved in introducing models into your business or have had a passing interest in economic affairs, you may have come across the term “Gini-coefficient”. This blog hopes to demystify the concept and give you a good deal of information on the statistical measurement. We answer:
Collection departments utilise diallers and collections management systems to improve their collections by segmenting the delinquent customers, prioritising them and applying a host of treatments. Whilst segmentation takes you so far, there are a host of other mathematical models that can be explored to improve what we call the “Collections Cascade”. Improvement in any step of the cascade can help improve the collections yield and there are a number of models that can be used. A few of them are listed below:
Bringing automation into the credit assessment process through credit scoring brings about significant benefits. Some of these benefits include:
This blog was originally published on 13 March 2019 and updated on 3 April 2019.
For a while, we have been running a blog series on cognitive biases and logical fallacies that data scientists should avoid. In this final blog on the subject, we look at some of the other logical fallacies and how they might crop up in data analytics.
For a while, we have been running a blog series on cognitive biases and logical fallacies that data scientists should avoid. In philosophy there are a host of informal logical fallacies – essentially errors in thinking – that crop up every day. In this series we have looked at the practice of data science to determine how these same fallacies also occur. Today we will be looking at fallacies and their manifestation in credit: The Monte-Carlo fallacy and the Hot-hand fallacy with some studies in the credit world.
As part of the group that was the second company worldwide to become IFRS9 compliant, IFRS9 has been at the forefront of what we do. We have assisted nearly 20 companies on their IFRS9 journey over the last two years. This blog forms part of a more extensive series on IFRS9. In this blog, we explore the administering of management overrides.
The Simpson's Paradox is a phenomenon in statistics illustrating how easy it is to misinterpret data. (Click to Tweet!) It occurs mainly in descriptive and diagnostic analytics (see our blog on the different types of analytics) where an analyst may jump to a conclusion driven by motivated reasoning and not by objectively assessing the evidence.
As part of our blog series on cognitive biases and logical fallacies that data scientists should avoid, today we address a prevalent logical fallacy: the "correlation proves causation" fallacy. Correlation due to causation is just one of the five main categories of causation, and this blog will look into each of the five.
A year ago, I published an article about motivated reasoning and how that can damage the data analytics process. It is part of a blog series on cognitive biases and logical fallacies that data analysts should avoid. Today I’d like to extend this conversation into a topical matter: p-hacking, also known as data fishing.
Business Rules Management Systems (BRMS's) are the Swiss-army knives of business software. Despite this, very few companies we work with are getting the most out of their decision engines. In this blog, we explore how BRMSs are used across the customer lifecycle.
Data science continues to be a hot topic in many large firms globally. 2017 saw data science subjects such as R vs. Python, deep learning, natural language, gamification, AI and machine learning being arguably the most topical.
As 2017 draws to a close, we reflect back on the year of credit. Some of the key themes that featured this year for us included:
As a company passionate about innovation we are regularly evaluating and re-evaluating knowledge – whether it’s our collective own, an employee’s or a client’s. Knowledge is a funny thing. Common sense might suggest that the more one learns about a subject the more confident one becomes. However, this is not entirely true, at least not in the beginning. The Dunning-Kruger Effect The Dunning-Kruger effect (DKE) is a cognitive bias that has been known for some time, but was only formalised in 1999 by two Cornell psychologists. It involves the seemingly contradictory idea that often those with little knowledge on a subject may come across as exceedingly confident about the subject. Conversely those with more knowledge may be less confident.
At Principa we’ve become quite passionate about Artificial Intelligence and Machine Learning. Recently quite a bit has been published in the press about how automated machines should be allowed to get. Most famously perhaps there have been the warnings from the likes of South African born Elon Musk and theoretical physicist Professor Stephen Hawking.
Most industries owe their levels of sophistication to the visionaries in their space. The South African credit industry is no different. Whilst the bureaux and the banks have played a significant role in developing the South African credit landscape, arguably the fashion retailers have also played a pioneering role in revolving credit. And so our vibrant industry owes much to the role of the fashion retailers. But how did it all begin?
Today we explore some of the frequently asked questions around mathematical optimisation. For the most part the questions are answered in the context of credit risk. However mathematical optimisation and operations research in general have many applications.
Although not a new concept, very few credit-granting organisations have deployed a true multi-bureau strategy in their organisation. It is, however, talked about fairly regularly, but often dismissed as “too hard” or “not important enough”. So why should you consider a multi-bureau strategy? What are the key considerations? How do you go about deploying a multi-bureau strategy? This blog series will address these questions.
In this blog post, I will be covering some of the highlights from day 2 of the Finovate Spring Conference. Whereas the previous review of day 1 mostly covered innovations that assist the loans origination process, in this blog I’ll cover some of the analytical offerings. I’ll also cover some of the other offerings including training, investment platforms, and bot-based technology.
We at Principa decided to attend Finovate Spring again this year (we were here 2 years ago) to learn about new trends, network with like-minded companies and to look for potential partnerships. In this post, I cover some of the highlights and more interesting topics from Day 1 of the industry's most exciting gathering of innovators in the fintech world.
Being a company that’s so passionate about software and analytical innovation means that we must constantly keep our ear on the ground for new trends and products around the world. There is hardly a better place for innovation than California’s Silicon Valley. Over the next couple of days (26th - 27th April), I'll be live-tweeting from this year's Finovate Spring conference in San Jose, California and summarising my take from a data analytics perspective in blog posts at the end of each day.
Use versus abuse of statistics can often be characterised by the analytical approach adopted to the problem at hand. In this blog post, which is part of a series on Logical Fallacies to avoid in Data Analysis, I’ll be focusing on defining the motivated reasoning logical fallacy and how to avoid it in data analysis.
The rise of Big Data, data science and predictive analytics to help solve real world problems is just an extension of science marching on. Science is humanity’s tool for better understanding the world. The tools that we use to build models, test hypotheses, look for trends to build value with our brand all derive directly from scientific principles. With these principles comes a myriad of obstacles. The obstacles are known to philosophers as “logical fallacies”, which I outlined in my previous post "The 7 Logical Fallacies to avoid in Data Analysis." In this blog post, we focus on the Texas Sharpshooter Fallacy and how to avoid it in your data analysis.
“Lies, damned lies and statistics” is the frequently quoted adage attributed to former British Prime Minister Benjamin Disraeli. The manipulation of data to fit a narrative is a very common occurrence from politics, economics to business and beyond.
If you’re involved in credit risk and existing customer marketing you’ll know that random numbers are frequently used when deploying different strategies. As strategies grow more complex and numerous, so the role of the random number grows more important. In this blog post, I’ll cover what randomisation is, why you should do it, when you should do it and what else to consider.
We've covered a few fundamentals and pitfalls of data analytics in our past blog posts. In this blog post, we focus on the four types of data analytics we encounter in data science: Descriptive, Diagnostic, Predictive and Prescriptive. Note: This blog post was published on the KDNuggets blog - Data Analytics and Machine Learning blog - in July 2017 and received the most reads and shares by their readers that month.
If I was to sum up our purpose at Principa, it would be “to help clients make informed decisions using data, analytics and software”. As information grows, so the opportunity to make better decisions increases. Data helps you understand your customer better. That’s our mantra. That’s our ethos. That’s why we are.
According to the World Bank, South Africans are the biggest borrowers in the world, with 86% of the population in debt.* And unfortunately, the National Credit Regulator, goes on to state that of the 20 million credit-active consumers in South Africa, 47% are in arrears on their accounts by three months or more, or had judgements against them, or had negative credit ratings on their credit record. ** As a credit risk professional, it is imperative to be able to segment your debtors by their propensity to roll and then apply the appropriate treatment by segment to minimise this risk and increase lift in your debt collection strategies.
It’s been almost 15 years since we saw the future of crime prevention in “Minority Report” – but today, we are beginning to see those then fictitious yet fantastical methods of predicting and preventing crime being implemented in various parts of the world. I’ll briefly mention three examples below of how analytics is already being used to prevent crime today before going into more detail on a fourth example: using analytics to prevent a criminal from re-offending.
In my previous post, we looked at the first “D” of the 3D approach of identifying and extracting value out of your transaction data: Determination. If you recall, I proposed a 3 step approach (the 3D approach) to realising value from a variety of large data sets: Determination –scour data sources to establish if and where there might be value Development – create models that will be developed for the decision areas where value was identified with data that is predictive of this predetermined outcome Deployment – implement and run the developed market
Introducing scoring into your business, means the adoption of a reliable and feature-rich scoring and segmentation engine that can calculate predictions or make recommendations that will minimise risk and maximise return on investment. This guide will identify critical features and nice-to-haves when assessing a scoring engine.
In the first part of our series on "Finding Value in Transaction Data" we explored a problem that is encountered by many organisations – how to identify and extract value out of the ever growing amount of data.
A scorecard is a mathematical model that is used to predict a certain outcome (for example the probability of default). The information used in a scorecard can vary, but common fields include demographic characteristics (e.g. age of applicant, number of dependants, time spent in current job) and credit bureau data (e.g. number of personal loans registered to applicant, worst arrears status on all account in the last 6 months).
The once flat region South of Johannesburg is littered with man-made yellow hillocks. These hills are a reminder of the history of Johannesburg – a city literally built on gold. The gold has been mined here since the rush of the 1880s. Much of the gold was extracted from the crushed rock and the left-overs transported by mules and later trucks to these locations. However, the mine-dumps (as they’re locally known) are yellow-coloured as a reminder that even after the original extract – allusive gold still exists in the waste piles.