The role of data analytics in strategic decision making in the financial sector

One of the most important sources of wealth in the 21st century is data. We live in the Information Age. Often the problem is not accessing them, since they are available, the problem is understanding and processing them in an agile and efficient way.

Table of Contents

Since Coinscrap Finance’s inception, our goal has been to simplify finance and make it accessible to everyone. Financial institutions struggle to analyze the banking transactions that are made by their customers, produce in-depth data analytics and leverage this data to offer a more personalized service.

The race to derive value from banking data has been an open challenge for some time now, and the PSD2 regulation has boosted it. The use of AI algorithms to automatically categorize and classify the short texts used in banking transactions makes it possible to obtain user data such as: income level, lifestyle, insurance, recurring payments, etc. This can improve the value proposition of banks and insurance companies.

Data analysis enables us to:

Óscar Barba

Co-founder & CTO of Coinscrap Finance

Automated classification of expenditure categories

In a survey of 500 global IT leaders by Pure Storage and Bredin, 86% of respondents indicated plans to increase data usage by 2023. Meanwhile, the analysis by Artem Mateush (et al.) makes it clear that automated payment classification has two approaches available: rule-based and machine learning-based. 

In the rule-based or regular expression approach, a set of guidelines is maintained to assign each payment record to a category. For example, a rule could be established that classifies all telecommunications-related payments as “Internet”. This approach is simple but can be complicated by the continuous updating of company data.

One alternative is to build a Machine Learning-based model from a set of labeled data. The approach would be to acquire your own or customers’ data and tag it with specialized equipment, which generates the knowledge base needed to train and build these solutions that simulate human reasoning and generate real-time output, similar to what skilled personnel would do. 

86% of respondents indicated plans to increase data usage by 2023.”

In addition, similar transactions from different customers may have different labels, known as the “noise” problem. Artificial Intelligence, and more specifically Machine Learning, allows the system to learn, spot patterns and be able to automate categorisation, although it is often complemented by some human supervision.

Offering customized financial planning tools

Currently, Deep Learning -a discipline within Machine Learning- is working to make the system’s learning autonomous and thus achieve the great challenge of simulating how the human brain learns. In order to carry out the automatic classification of text, the algorithm must be applied to each transaction.

One of the most widely used is the support vector model, which is applied to information retrieval systems. In addition, other forms of text representation exist such as graphs, n-grams, logical representations, etc. Automated text classification has become popular due to the large amount of digital content available.

Thanks to these advances and the digitalisation of finance, it is now possible to show bank users how, where and when they are spending their money. With the current economic situation, it is necessary for banks to position themselves as an ally and offer tools with which their customers can regain control of their financial situation and make better decisions. 

Create a Expense Manager with data analytics

A good example is a expense manager or PFM, modules that draw on data analytics to show an overview of consumption habits, recurring income and insurance due dates, among others. PSD2 has also contributed to these tools, making it possible to aggregate accounts from several banks and thus enrich the 360º image of the user’s finances

In this article you can learn more about what a Expense Manager is and what it is used for.

All their banking movements are shown, regardless of the bank where they are generated and, through these AI technologies, they are homogenized in the classification. There are also micro-savings tools, designed for people to create goals and increase the amount in their piggy banks with little friction and with a digital and gamified experience.

These developments are particularly useful in environments such as the Spanish market, where the population has a lack of financial literacy. The survey cited in, on Financial Competences, carried out by the Bank of Spain and the National Securities Market Commission (CNMV), placed Spaniards below the average of 40 OECD countries

Its findings reflect people’s difficulties in budgeting and a low level of understanding of financial terms. The same applies to saving and the purchase of banking and insurance products, which end up jeopardizing people’s financial health.

Hyper customisation of products and services

As mentioned, Deep Learning is a branch of Machine Learning that is commonly used in the recognition of complex patterns in unstructured information. Using neural network architectures, the use of transformers is currently gaining presence. It can be applied to any task, but incorporates the possibility of working with non-tabular input data.

In the financial sector, data analytics and data categorization, the probabilities are enormous., from creating user typologies and making recommendations to offering personalized services to customers based on their interests in real time. All thanks to AI. To give an example, we process millions of banking transactions daily for Banco Santander.

New call-to-action

We must not forget that, depending on the different groups of users obtained by typology, intelligent alerts or insights can be launched, according to the situation and events detected for each customer, in order to offer services and products in a totally individualized way. Adapting messages with these technologies is vital to avoid being left behind in the innovation race.

About the Autor

Óscar Barba is co-founder and CTO of Coinscrap Finance. He is an expert Scrum Manager with more than 6 years of experience in the collection and semantic data analytics in the financial sector, classification of bank transactions, deep learning applied to stock market sentiment analysis systems and the measurement of the carbon footprint associated with transactional data. 

With extensive experience in the banking and insurance sector, Óscar is finishing his PhD in Information Technology right now. He is an Engineer and Master in Computer Engineering from the University of Vigo, Master in Electronic Commerce from the University of Salamanca, Scrum Manager and Project Management Certificate from the CNTG, SOA Architecture and Web Services Certificate from the University of Salamanca and more.

Sign up for our newsletter and get our top stories delivered straight to your inbox