Scaling Your Institution’s Screening with GOST

One of the critical challenges to continuously screening full populations is scalability, or the ability to quickly process data on hundreds of thousands or millions of entities daily. A large bank, public institution, or any organization that needs to screen large numbers of people quickly, cannot sustain workflows at the necessary scale without getting overwhelmed by the alert yield. Institutions often have to make hard choices, including screening subsets of the population rather than the entirety of it.

Matthew Troke/Shutterstock.com

First, what is scaling? If an institution has a small population, their professionals can manually run negative news searches to screen effectively. The workforce is a match for the labor required for the task. However, as the population needing to be screened increases, technology is required to achieve scale. This makes sense: a small family can garden a few acres to provide food for sustenance, but if they acquire more - say, thirty or fifty additional acres - they will need to invest in technology, like a tractor, to scale their efforts. In this, agriculture and screening are the same. Technology helps to complete a task that is too large for humans to do alone but is ultimately to their benefit.

Currently, industry solutions for large-scale needs include tools that support list-based matching of names on lists like sanctions and PEP screening. This data, however, is highly structured and generally static. This is not helpful for information that is subject to change, or unstructured information. In addition, the ability to screen over 1,000,000 searches a day means that institutions can screen and continuously vet entire populations quickly and meaningfully.

Let’s explore two reasons why this is incredibly important.

First, lists of structured data are not the best solutions for finding negative news in a dynamic world. People are on a list, or they are not: but that does not mean that they have no risk-indicative behavior helpful to the screening process. Lives are lived online, and those lives are not always recorded on a list. Sometimes they are, but even then lists can take time to be updated in local law enforcement databases.

We can illustrate this with a story: someone commits a crime that was recorded in a local newspaper. That same crime was recorded by police, but the local database is only updated once every month. In the meantime, an institution screens the person’s name, unaware of his criminal behavior. Without an updated database, the institution does not know that this person should not be approved. But, if the institution had access to unstructured data beyond lists, they would have seen the local newspaper report and taken a different action.

This story is fictional but has been mirrored in real life. Lists are great sources of information, but they are static and often not reflective of adverse media. GOST allows institutions to use both without having to compromise on data source or population subset.

The second reason that GOST’s scalability is important is the ability to continuously monitor a population. One million searches per day means that GOST doesn’t just scale to the volume requirements of an institution, but also to the frequency requirements. A bank that used to refresh a customer population every two years can now do it every quarter - ensuring the most updated information and risk assessments. An investment management company that monitors sources of reputational and financial stress can choose to refresh data weekly, rather than annually - making their assessments even more precise.

With scalability, GOST can identify potential sources of risk as fast as information changes, empowering analysts to follow up on the highest value leads - on an ongoing basis.