Smart Banking versus Financial Pathology

- Innovative Analytics Herald A New Era

In our last two issues we highlighted how data (analytics) is the key factor in setting you apart from your competition as a financial institution. We also took a deep dive into the four most formative phenomena in banking nowadays and how two of them, democratization and commoditization, inhibit catalytic power in order to unfold hidden potentials, boosting and accelerating the competitive advantage of a modern Private Bank.

Today, after having laid this foundation of reasoning, we would like to switch gears again: we will take a closer look at some of the most innovative Understand Your Customer (UYC) techniques and practical examples applied in modern Private Banking.

These techniques constitute the difference between 'modern smart banking', based on early-warning, forward-looking, real-time analytics, which reflect and connect to outside reality, on the one hand, and 'financial pathology', which represents a delayed, backward-looking analysis, occurring after the fact and mostly based on scholarly theory, with only a limited, 'theoretical' connection to reality, on the other hand.

The mother of all holistic UYC techniques: Dynamic Profiling

The term UYC itself implies a thorough understanding of a myriad of dynamic processes and the capacity to connect the dots - continuously. Although we focus on the clients themselves, the mentioned dynamic processes comprise and imply understanding of a very complex and highly dynamic client environment as well.

The backbone of all holistic UYC techniques is formed by 'Dynamic Risk & Opportunity-Based Profiling'. Despite the unjustified negative connotation of the term 'profiling', which has grown historically, this concept remains one of the most powerful and fundamental techniques allowing for the creation a dynamic 360° analytical view of clients and their steadily evolving and influencing environment. Every new incoming or outgoing transaction, business relationship, conversation with the client advisor, or newly collected counterparty artefact has a direct influence on the continuous (hence dynamic) recalculation and updates of the client profiling. However, it might come as a surprise to many practitioners that the technique of Dynamic Profiling, does not necessarily involve any advanced ML or AI techniques, despite its important and indispensable role. First and foremost, it is built on good old statistics.

More advanced analytical UYC techniques with usage of ML and AI such as 'Customer Behavioural or Sentiment Analysis' build on the foundation of 'Dynamic Profiling'.

The limits of UYC techniques in PB – and five solutions

As already mentioned in our first issue, relying on 'Transaction Monitoring' in Private Banking is unfortunately quite insufficient compared to retail banking, where transactions and hence client insights are abundant. This means we need to focus on additional (re-)sources.

Let us have a look at five solutions and how they show up in daily business – let us therefore meet Ms Miller and her client Mr Mayer:

Ms Miller

Relationship manager at a Private Bank

Mr Mayer

Private Banking client

1. 'Speech and Sentiment Analysis'. Every verbal interaction between the client and the client advisor is transformed into text or even structured data. This is then analysed in the context of a variety of questions, such as 'Sentiment Analysis', 'Customer Attrition', 'Cross & Up-Selling' along with any compliance- and regulation-driven topics, and finally put in the context of the constantly updated 'Dynamic Risk and Opportunity-Based Profiling' of the client.

During a call with Mr. Mayer, based on the ongoing Speech and Sentiment Analysis, Ms Miller is notified about warning signals pointing to an increased dissatisfaction. Put in the context of his Dynamic Risk & Opportunity-Based Profiling, the system registers an increase in the probability that Mr. Mayer may leave the bank.

2. 'Dynamic Portfolio Analysis' in combination with 'Predictive Financial Market Analysis' are two other UYC techniques that make the 360-degree view even more complete. Based on the constantly updated client profile - the understanding of who the client really is and what he/she really needs - and on the ongoing dynamic analysis of the financial markets, the client advisor can offer the best, most suitable and customized services, including those distributed through the new digital, self-service PB channels.

Based on the Dynamic Portfolio Analysis and the Financial Market Forecast, Ms Miller is notified about the continuously increasing tendency of Mr. Mayer’s portfolio towards emerging markets over the last few weeks. In combination with the market forecast for the next three months, Ms Miller sees that this exposure could lead to a significant deviation from his investment objectives and actual investor profile.

3. Another very important UYC technique is the 'Ongoing Model Optimization, Validation and Calibration'. This technique is basically THE manifestation of the key term 'Dynamic' in its essence, irrespective of the type of analysis we are about to conduct, be it Fraud Detection & Prevention, Customer Attrition, Cross & Up-Selling, or any other. Thanks to this ground-breaking concept, it is possible for us to know which factors play a significant role in the evaluation of the client behaviour right now.

Based on the overnight model 'Optimization, Validation and Calibration' of the Customer Attrition, the Dynamic Portfolio Analysis, the Financial Market Forecast model, as well as the Customer Segmentation Analysis, the middle office of the bank notices a significant shift in the relevance of some of the model factors related to clients with a similar client profile to Mr. Mayer.

4. The aforementioned knowledge about all the relevant factors allows us to breathe new life into a relatively old and well-known technique: Transition Matrices. Thanks to the inclusion of the 'Dynamic' factor, the whole concept has undergone a complete remake and rollover and is now spoken of as 'Dynamic Migration Matrices'.

In contrast to Migration Matrices, where we simulate and build scenarios mostly on generic, e.g. macroeconomic, factors, with Dynamic Migration Matrices we can dynamically simulate using precisely identified, relevant, client- and environment-specific factors, which constantly undergo changes and evolution.

Based on the shift in relevance of model factors, Ms Miller runs several simulations and stress tests on these new factors to see how the forecasted and likely development of these factors will influence the portfolio performance of Mr. Mayer, his level of satisfaction and the likelihood that Mr. Mayer will transfer his account to a competitor. Ms Miller addresses these points during her next call with Mr. Mayer and offers him several alternatives to handle and mitigate these portfolio risks. Moreover, she offers him several new communication methods, which should significantly increase Mr. Mayer’s level of satisfaction and strengthen the client relationship.

In the meantime, with the help of 'Speech & Sentiment Analysis', back-office and internal audits have identified potential for improvements regarding their client advisor’s compliance with internal cross-border and tax evasion guidelines. Therefore, additional training sessions for client advisors have been set up.

Smart Banking or Financial Pathology?

The application of all of these innovative techniques allows for a precise, ongoing, forward- looking and well-founded UYC analysis and at the end of the day for adequate and timely decisions, actions and measures, no matter which specific UYC techniques or combination thereof a bank will go for.

Knowing exactly what constitutes, shapes und influences our clients and their behaviour with the help of these modern analytic tools supports us in making the best and well-founded decisions, in offering the best possible service to them, protecting their and our integrity and finally, just being smart, forward-looking bankers and economists - instead of backward-looking financial pathologists.

Contact us



About the author

Nikolai Tsenov is Product Manager Analytics at Finnova and has been with the company since 2015. He is responsible for the conceptual design and development of the Finnova Analytical Framework (FAF), a unified and holistic analytical platform that fulfils all of a modern financial institution's analytical needs. Nikolai has over 20 years of experience in the fields of data analytics, compliance and risk management.

With his main focus on banking and finance, he has worked on and managed multiple international consulting and implementation projects in the fields of fraud detection and prevention, AML and sanction screening, transaction monitoring and behavioural analysis, aCRM, Basel II/III, cross-border compliance, and product and customer suitability. He holds Master's degrees in International Economics and Finance from the universities of Innsbruck and Barcelona and is the winner of the 'Banking IT Innovation Award 2016' from the University of St. Gallen and the 'Fintech Breakthrough Award' in the category 'Best Predictive Analytics Platform 2020' with his analytical framework concept FAF.