Natural Language Generation in the Financial Sector – Practical Experience

Reading Time: 2 minutes

Natural Language Generation in the Financial Sector – Practical Experience

It is always exciting to see in how many sectors NLG is used successfully. Today Mario Berger, Director at Deloitte, reports on his experiences with the AX Semantics software.

Hello Mario, please introduce yourself to our readers.

I have been in the financial services industry for more than 20 years and have been responsible for the introduction of Natural Language Generation at Deloitte. I started my career at a major German credit institution before moving to banking and insurance consulting. My focus is on reporting and regulatory reporting, the implementation of regulatory requirements for reporting, and the digitization of all these topics. I also moderate various formats such as strategy, innovation, and business workshops and manage projects of varying scope and content.

When did you start to focus on NLG (Natural Language Generation)?

It all started with my first projects because process efficiency and optimization were already at the top of the agenda in regulatory reporting. For example, when implementing regulatory requirements, we have always considered how to set up repetitive or pattern-based parts of reporting using rule-based systems and curated templates. Since early 2018, the Deloitte Network for Language Technology has been pushing further development of “language technology.” Together with research institutions, universities, and solution providers, we are working to ensure that technology transfer into today’s process world is successful. And we also are looking at ways to increase the acceptance of NLG in the public sector and in companies. 


What potential does Deloitte see in automated text generation?

If one refers to the creation of certain recurring reports, various projects have already shown that the effort can be reduced by up to 80%. In our estimation, projects in this environment are significantly more successful if parts or even the entire “path” of the process is automated – i.e. data management, for example, using Natural Language Processing. We consider the fundamental potential to be very high. 

We are also currently working on a study that deals with experience in NLG handling and the findings regarding cost and process efficiency. 

What concrete projects has Deloitte realized with NLG?

Our focus is on the reporting of established processes. In addition to the successful implementation of risk and compliance reports, we examine whether NLG can also be used efficiently in the area of finance or balance sheet preparation. There is also considerable potential in the area of auditing, and with DNAV we already offer a digital solution for Smart Fund Audit Analytics. 

Why did you choose to work with AX Semantics?

The Text Engine of AX Semantics fulfills all essential aspects for the generation of the mentioned reports from our point of view. Quality and applicability of the solution are very high, which was also confirmed by the Forrester report from October 2018. And last but not least, the operational cooperation between the teams is excellent, and we get on really well with the AX Management Team. 

About Deloitte

Deloitte provides auditing, risk advisory, tax advisory, financial advisory, and consulting services to companies and institutions from all sectors of the economy; in Germany, legal advice is provided by Deloitte Legal. With a worldwide network of member companies in more than 150 countries, Deloitte combines outstanding competence with first-class services and supports clients in solving their complex business challenges. Making an impact that matters – for Deloitte’s approximately 312,000 employees, this is not only a common mission but also an individual challenge. Deloitte refers to Deloitte Touche Tohmatsu Limited (‘DTTL’). A more detailed description of DTTL and its member companies can be found at


You can find more testimonials for AX Semantics in the interviews of Bernd Vermaaten from, Stephan Schütrumpf from Creditreform and Harry Olfert from Schäfer Shop.