In this article, I delve into a fundamental topic for boosting customer retention and engagement: the analysis of transactional data. Thanks to our research and the training of our proprietary AI engine, COCO, we help banks gain deep insight into the circumstances, concerns, and needs of their customers.
Óscar Barba
Co-founder & CTO of Coinscrap Finance
Turning global trends into real business opportunities
As Deloitte points out, communication between businesses and customers will soon be personalized to levels that seemed unimaginable just months ago. “For example, Graybar is testing AI in sales and customer service (…) by embedding artificial intelligence into systems that help account managers uncover cross-selling and upselling opportunities.”
We see this every day –data-driven models are shaping the future of customer loyalty. Our technology can analyze behavior patterns, preferences, and context to predict each customer’s next step. For example, COCO can identify when a user is likely to need a new financial product, deliver personalized offers at the right moment, or automate routine tasks to simplify daily banking.
It’s no longer just about answering questions or offering loyalty points. It’s about creating an ecosystem where the brand anticipates users’ needs and offers solutions before they even ask. The data is there. The infrastructure exists. The question is: why wait to transform your customers’ financial experience?
Use cases for the financial industry
Leading bank marketing departments are already using tools capable of processing millions of banking transactions per second, uncovering deep and actionable knowledge about each user: from spending forecasts, consumption habits, and personal preferences, to financial needs, risk behaviors, and seasonal payment patterns.
These technologies can detect key life events. For instance, spending on daycare, pharmacies, and baby product stores may indicate the arrival of a newborn. Similarly, domain purchases, office rentals, or notary fees may suggest someone has started a business.
Both scenarios signal new responsibilities. Being able to identify them in real time enables banks to offer relevant products –such as life insurance– at the exact moment a customer might be thinking: “What if something happens to me?”
Such levels of personalization were once out of reach for many companies. Today, strategic collaboration between banks and fintechs is democratizing access to these capabilities.
Bank-Fintech alliances: the key to innovation without massive investment
Banks that adopt these tools gain a distinct edge in an increasingly competitive industry. Neobanks, big tech companies, and new financial players are rapidly reducing the traditional banks’ market share. That’s why many are turning to fintech partnerships to implement innovative solutions –without the cost and complexity of building them from scratch.
The alliance with fintechs has proven to be the key to enhancing banks’ ability to create highly segmented products and services. These collaborations allow banks to leverage the agility, technology, creativity, and customer-centric approach of fintechs –without bearing the costs of in-house development.
I’d like to show you how all this theory comes to life in a practical case: our team has developed a project supported by the Centre for the Development of Industrial Technology (CDTI), a public business entity under Spain’s Ministry of Science, Innovation, and Universities.
💡 Did you know that…
“Our algorithm can help you determine which customer is ready for a new financial product.”
Success story: our project for CDTI Innovation
Coinscrap Finance’s project, “Hyperpersonalized Finance Intelligence: Empowering Users Through AI-Driven PFM”, highlights the importance of algorithmic models capable of detecting behavioral patterns and extracting actionable insights from customer purchase habits and financial preferences.
This work explores:
- Clustering users based on similar transaction behavior
- Detecting relevant life events
- Product recommendation models
- Predicting future product adoption
One of the key breakthroughs from our research –already helping major banks strengthen customer relationships– is the ability to hyperpersonalize financial product recommendations. Our intelligent modules detect opportunities for institutions, such as identifying when a customer might benefit from refinancing debt.
Let’s look at this example in more detail: After analyzing training and test datasets, we pre-process the information to build a model that can assess the suitability of specific customer segments. The goal is to detect those who might benefit from a debt refinancing insurance product based on their financial behavior.
Debt refinancing insurance:
- Cluster 1: Individuals with high debt-to-income ratios and near-maxed credit utilization.
- Cluster 2: Users with unstable or highly variable income.
- Cluster 0: All other users (typically with lower saving capacity).
Our model selects indicators based on the user’s financial level, and the project also takes into account other variables, such as the information a human agent would need to assess suitability. Subsequently, specific aspects of the debt and current interest rates would also be included.
In this way, the algorithm chooses the clients who are most likely to accept this product and benefit from the offer the bank makes for their specific case. This is how financial institutions increase their business volume: by creating upselling and cross-selling opportunities within their existing customer base.
Final thoughts: the financial industry must adapt to the real impact of AI
Our research demonstrates that it’s now possible to anticipate user needs, recommend appropriate financial products, and deliver memorable experiences by detecting behavior patterns and enabling dynamic segmentation.
Financial institutions that adopt this technology with a strategic mindset will be better equipped to compete in an increasingly demanding environment. Hyperpersonalization is no longer a promise –it’s a real, measurable competitive advantage.
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 analysis of data 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.
Doctor in Information Technologies from the University of Vigo, he holds a Master’s degree in Computer Engineering, a Master’s in E-Commerce from the University of Salamanca, a Scrum Manager Certificate, and Project Management certification from CNTG, as well as a Certificate in SOA Architecture and Web Services from the University of Salamanca. He has recently obtained the ITIL Fundamentals certification, which recognizes best practices in IT service management.
