AI and the automation of credit decisions, the...

Bazadise has implemented various technologies such as Artificial Intelligence (AI) and Lambda from Amazon Web Services (being the success story of the Amazon cloud), to allow risk and credit teams to adapt their credit models autonomously and simply, without the need for a technology team, since the tool is very intuitive and has a graphical editor to implement rules.

In recent years, in Latin America, a great boost in financial inclusion was reflected, which was presented, in part, thanks to the significant contribution of technologies within the sector, which influenced the democratization of access to different products.

In Colombia, for example, there were more than 32.7 million adults, with at least one savings or credit product, which meant that 89 out of every 100 Colombians had decided to join the financial system, according to a report on financial inclusion carried out by the Banking of Opportunities.

Faced with this panorama, the great work carried out by fintech should be highlighted, since they are the ones that have implemented technologies to transform traditional financial products and services, taking into account sectors of the population such as young people, peasants and even women, which for a long time were not taken into account by more conservative entities.

And within this ecosystem, at the end of 2020 came uFlow, the North American company that developed a 100% cloud credit decision engine that allows the risk and credit teams of any financial institution, whether banks, fintechs or retailers, to improve their operation flexibly and without depending on the IT area when making changes to the rules or models that automate the granting of loans to people and companies.

In this way, and through constant innovation processes, Bazadise has implemented various technologies such as Artificial Intelligence (AI) and Lambda from Amazon Web Services (being the success story of the Amazon cloud), to allow risk and credit teams to adapt their credit models autonomously and simply, without the need for a technology team, since the tool is very intuitive and has a graphic editor to implement rules.

Bazadise was born for risk and credit teams to self-manage their lending policies and improve their operation, since they do not depend on the IT area to make changes in the rules when granting loans.

Thanks to their constant evolution and the application of new technologies, they have already managed to reach different Latin American countries such as Colombia, Argentina, Peru, Chile and Panama. And at the end of 2021, the number of transactions generated through this tool increased by 149% compared to 2020, after going from 2.4 million operations to more than 6 million, projecting 2022 with 12 million transactions.

It is also very important to note that your decision engine takes only 0.1 seconds to analyze the data of each user within the credit bureaus, see the guidelines implemented by the financial institution and resolve each transaction; allowing people to know if their loan is rejected or approved, how much can be granted and in how many installments they would have to pay it.

According to a study carried out by the Banking of Opportunities, it was found that the main barriers to access to credit in Colombia are: self-exclusion (64%), requirements (20%), high costs (10%) and negative reports in credit institutions (5%). These facts are partly factors that were generated due to the imaginary that traditional financial institutions built for years.

Faced with this, the director of Bazadise believes that providing technological solutions generates a change and also helps to renew the banking sector. "We promote financial inclusion by streamlining and automating decision-making, incorporating multiple sources of data, and allowing financial institutions to have more information about people who apply for credit in a simpler way, which generates them to make better decisions avoiding human error, and without depending on other areas to improve their offer every day."

In Colombia, for example, this decision engine is currently connected to nine different data sources, among which are: TransUnion, Experian, Mareigua, Arus or Asofondos, among others.