top of page
We are inundated with data; we're practically drowning in it. Financial regulators and supervisory authorities are exacerbating this trend by continually expanding data and reporting requirements for assessing and evaluating the creditworthiness of market participants, their ESG impact, and profitability. Analyses of counterparties must encompass all available datasets comprehensively. For banks, auditors, strategy consultants, or finance departments of corporations, this demands legions of financial analysts dedicated to establishing a unified database before they can embark on their actual work—analysis.
Collecting data as swiftly as a dragonfly in flight
Why is this the case? Because all the data – figures and facts – are not uniformly structured. An outsider might assume that there are standards to be followed in the financial industry. However, this is not the case. Each institution, bank, or company discloses their data differently, sometimes using different terminology and often employing different calculation logic, not to mention incomparable layouts. This makes it laborious for an analyst to compile the numbers of a company from different sources in such a way that they fit into the evaluation scheme of their own analysis method, rating model, and so on. In the analysis of banks and insurance companies, for example, this often requires an entire working day for preparation for just one counterparty. This is laborious but unavoidable. While one may argue that ready-made databases exist, they are expensive and may not meet the requirements of the particular analyst.
We believe that modern technology, especially AI, can tackle this challenge. Our aim is not to replace analysts and human judgment, but rather to significantly accelerate and streamline the process of data gathering and preparation through the use of intelligent solutions.
Collecting data as swiftly as a dragonfly in flight
However, we also encounter significant technical challenges that may appear simpler than they truly are. Firstly, the numbers must be extracted from annual reports or other disclosures, typically provided in PDF format. As detailed in our application, even this initial step lacks a satisfactory technical solution. Just like other seemingly simple steps of data processing:
1.Data extraction from PDF documents;
2.Correct interpretation of different table formats;
3.Proper assignment and labeling of extracted figures
4.Representation in a universally valid data structure;
5.Aggregation to the key figures needed by the analyst and evaluation method;
6.Plausibility check of the results.
If we succeed in developing a technical solution that can automate a substantial portion of these steps, the potential scope of application is enormous. Currently, these tasks are predominantly executed manually by thousands of individuals, not limited to the finance sector alone. This presents a significant market potential. However, creating a simple and quick Minimum Viable Product is not feasible.
Collecting data as swiftly as a dragonfly in flight
Why do we believe we can solve this? Because we bundle the necessary skills in one team and can bring together experience and competences from all over Europe: mathematical and technical proficiency along with extensive management experience. Our international team has a track record of successful development and implementation of complex projects, as well as the required scientific background.
Why is this important for Austria? Because Vienna, as one of the essential Central European financial centers, can benefit greatly from such groundbreaking technical innovation.
Learn about the founding team
Do you want to invest?
bottom of page