How a real estate developer optimized credit granting with Machine Learning and reduced delinquency by 46%.
- Marcelo Firpo
- 5 days ago
- 5 min read
Standardization in credit analysis Automation with predictive models Scalability. with data-driven decisions
AI-generated summary:
Seeking greater precision and speed in credit granting, a large Brazilian real estate developer adopted a machine learning-based solution developed by BlueMetrics. The predictive model was trained using the company's own historical data, considering variables such as income, marital status, and number of children, to automatically classify the default risk of new applicants. The solution, integrated with AI and supported by AWS technologies, eliminated subjectivity in the process, reduced rework between departments, and enabled faster and more informed decisions. The result? A potential 46% reduction in default rates.
Overview
In a highly competitive real estate market, ensuring that credit is granted to the right clients can make all the difference. This was precisely the challenge that one of Brazil's largest real estate developers faced and overcame by using artificial intelligence applied to credit analysis.
Our client, a real estate company that sells residential units, also operates as a financing provider, granting direct credit to buyers. With the increase in sales volume and the growth of operations, the need to revise the credit granting model became evident, as it was previously poorly standardized and highly dependent on the subjective assessment of analysts.
“Working on problems like this is extremely motivating for us because they are strategic for the business and allow us to apply our expertise in a practical and measurable way,” says Luciano Rocha, commercial director of BlueMetrics. “Furthermore, we know that a well-organized data structure makes all the difference when developing AI solutions that truly deliver value, and data expertise is precisely one of our greatest differentiators.”
Market context:
High competitiveness in the real estate sector.
Increasing volume of loan applications
The need for quick and assertive decisions.
High risk of default.
Manual processes that are susceptible to human error.
Problem: How can we optimize customer service and provide accurate answers regarding financing?
As we saw above, in addition to selling residential properties, the developer also offers its own financing, which increases its profit margin but also increases financial risk. The credit analysis process was conducted manually, based on criteria that varied among analysts, including even subjective factors such as their mood that day or pressure to meet targets.
This lack of standardization generated inefficiency, conflicts between the commercial and financial areas, and hindered risk control. In a market with high default rates and short decision-making deadlines, the company urgently needed a more objective, reliable, and scalable approach to assess credit risk consistently and quickly.
“We have seen this type of challenge repeated in various sectors: a lot of data available, but little strategic use,” highlights Luciano Rocha. “That’s where our experience helps companies transform all this potential into concrete initiatives.”
Operational limitations:
Manual and time-consuming credit analysis process
Dependence on the individual judgment of analysts.
Lack of standardization in decisions
Conflicts between areas due to differences in assessments.
Business limitations:
High risk of default.
Difficulty in scaling the operation safely.
Impact on customer experience due to delays.
Critical decisions influenced by subjective factors.
Technological limitations:
Lack of an automated model for risk analysis.
Lack of integration between data and company departments.
Lack of a consolidated history of previous decisions.
Low capacity to generate insights from available data.
The solution: Machine Learning in credit analysis
To overcome this challenge and generate real value, BlueMetrics developed a machine learning-based classification model capable of predicting, based on variables such as income, marital status, and number of children, whether a loan applicant would have a higher or lower propensity to default.
The model was trained using the company's own historical data and integrated with an artificial intelligence agent that performs the query in real time. As soon as a new credit request is received, the system automatically assesses the risk, generating a risk score to support the analysts' decision.
Important: ultimate control remains in the hands of the human team, but now with the support of objective and consistent data.
The architecture was built with scalable AWS technologies, such as Amazon SageMaker, ensuring performance, reliability, and flexibility for operational growth. Gabriel Casara, CGO of BlueMetrics, adds: “We have already delivered around 200 data and AI projects to more than 90 clients in Brazil, the United States, and Latin America. And it is precisely this track record that allows us to offer speed in solutions, security in delivery, and a total focus on results.”
Immediate benefits:
Reducing the workload of analysts
Standardization of analyses and elimination of subjectivity.
Automated customer service and credit analysis 24/7
Greater agility and assertiveness in decision-making.
Strategic gain:
Potential reduction in default rates by up to 46%
Structured database for future marketing and credit initiatives.
Supporting decision-making with reliable historical data and forecasts.
Possibility of scaling the operation safely and efficiently.
Key features of the solution:
Integration with AI agent for real-time responses.
Total transparency and control on the part of the analysts.
Implementation using robust and scalable AWS technologies.
Model trained with real data from the business itself.
Want to see GenAI and Machine Learning solutions
making a difference in your company?
Results:
The solution achieved 92% accuracy in classifying good payers, making the process more reliable and significantly reducing rework between departments. Standardization brought clarity, reduced internal conflicts, and improved operational efficiency.
According to simulations using historical data, the company achieved a potential reduction of up to 46% in delinquency, in addition to a significant gain in the speed of credit decisions.
Furthermore, the model began generating valuable insights for the marketing team, which started targeting campaigns based on customer profiles most likely to pay on time. This created a virtuous cycle of efficiency and prevention, with a direct impact on the profitability of the operation .
Technologies used
The solution was designed using several AWS technologies, including:
AWS Services
Sagemaker
S3
Lambda
DynamoDB
API Gateway
Languages, Libraries and Frameworks
Python
Conclusion:
This case demonstrates how the application of artificial intelligence in the real estate sector can go far beyond automation. By bringing predictability, agility, and intelligence to the credit granting process, the company was able to reduce risks, make more strategic decisions, and scale its operations safely.
More than just a one-off improvement, the project represented a leap in the organization's analytical maturity. "It's very gratifying to see a solution generate real and immediate value for the client, solving a concrete problem with a direct impact on results," concludes Luciano Rocha. "That's what we strive for in every project we deliver."
How about creating a case study like this for your company? Let's schedule a call?
Learn about some Use Cases .
About BlueMetrics
BlueMetrics was founded in 2016 and has already delivered over 160 successful projects in the areas of Data & Analytics, GenAI, and Machine Learning for more than 70 companies in the United States, Brazil, Argentina, Colombia, and Mexico. It has its own methodology and a multidisciplinary team focused on delivering solutions to real-world business challenges.
Comments