Many financial institutions that are required to implement Know Your Client (KYC) and Anti-Money Laundering (AML) measures view them as a burden and a liability. But in fact, these practices are crucial for the ongoing smooth operations of a financial institution and, globally, for maintaining stable economies, free of ‘dirty’ untaxed money and criminal activities. The increasing degree of digitalization has opened the opportunity for data analytics to improve KYC and AML monitoring. The target is to bring KYC profiles into alignment with the corresponding client’s transactional behaviour, automatically, and at scale. In this talk we present two real-world examples of significant KYC quality boost fuelled by data analytics, which we implemented and tested with a client. The first example show-cases how fine-tuning of deep learning NLP (Natural Language Processing) models can lead to accurate transaction destination detection. The second use case demonstrates how account clustering based on aggregated transaction behaviour helps efficiently detecting account mis-usage. These examples illustrate how tailored data analytics solutions efficiently detect KYC problems that are extremely difficult and time-consuming to find manually or by applying simple rules.