FintechAdvanced Analytics & AI
Credit Documentation Project
We solve the problematic of credit documentation applications with a high probability of not completing necessary requirements that cannot be detected in time.
- Industry
- Fintech
- Solution
- Advanced Analytics & AI
01 / 03
The challenge
- Credit applications with a high probability of not completing the necessary documentation requirements cannot be detected in time, which leads to the sale of a financial product not being completed.
02 / 03
What Tekne did
Objective
Establish with an acceptable degree of certainty, considering the history reflected in the database, the probability that a credit application file is on the way to completing 80% progress.
Models
Construction and implementation of a Machine Learning solution for:
- Identify the key variables
- Predict which cases are more likely to not complete the requirements for the credit application
- Profile customers to offer other financial products
03 / 03
Results
- It was possible to identify in time the files with a low probability of progress, allowing better management of the cases
- The variables that most influence the credit application process were defined
- An increase in successfully completed applications was recorded
Initial Situation
Without model
26%
of requests did not reach 80% progress
Final Situation
With model
97.3%
effectiveness in predicting cases with a low probability of progress
Let's talk
Book a meeting