1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Fintech
Loan management system with 4x efficiency boost
Re-engineering the solution’s architecture and adding new logic
Dedicated team
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Fintech
Loan management system with 4x efficiency boost
Re-engineering the solution’s architecture and adding new logic
Dedicated team
Who we worked with
Client overview
Estonian loan management software provider
Who we worked with
Client overview
Estonian loan management software provider
What we did
Project overview
- Problem
Our client had already built a strong foundation with their core lending platform, but one critical piece was holding them back: the credit scoring module. At first glance, the system worked — it could evaluate borrowers and help decide whether to issue loans.
But beneath the surface, there was a problem.
Each new financial institution had its own scoring criteria. Adjusting the solution to them required developers to make changes directly in the code. That meant they were constantly pulled into repetitive tasks, tweaking rules for age, salary, loan history, or expenses, instead of focusing on innovation.
On top of that, monolithic architecture didn’t allow to introduce changes on the go.
This rigidity created a bottleneck. For our client’s customers, it meant slower onboarding and limited flexibility. For our client themselves, it meant higher costs, wasted resources, and a product that couldn’t scale as quickly as the market demanded.
- Our solution
We added the new logic to conveniently tweak credit scoring rules directly from the tool’s interface. This allowed our client’s staff to adapt them to each new customer without involving developers.
We also extended the app’s functionality to process smaller loans automatically. It collected various data on applicants from their profiles and other sources and applied a grading algorithm. For example:
1. Employment stability:
– Full‑time job with 3+ years at current employer → +15 points
– Frequent job changes in the past 2 years → –10 points
2. Savings & assets:
– Maintains a savings account with consistent deposits → +10 points
– No savings or assets → –10 points
3. Payment behavior:
– No late payments in the past 12 months → +15 points
– Multiple late payments in the past year → –20 points
4. Credit utilization:
– Uses less than 30% of available credit → +10 points
– Maxed out credit cards or overdrafts → –15 points
(the reference doesn’t illustrate the actual scoring criteria and algorithm)
Summarizing results, the tool worked out the final grade and made a decision on approving/rejecting the loan. Riskier or more complex loans still required manual review after the automatic assessment.
Pursuing our vision of tech efficiency, we also re‑engineered the platform’s architecture. We broke the monolithic application into microservices that made the system easier to support, customize, and connect with other tools.
- Workflow
The development team operated fully remotely, following Agile methodology with structured sprints and bi-weekly demos to ensure continuous feedback and alignment. Transparent communication and progress tracking were maintained through Jira for task management and Slack for daily collaboration.
What we did
Project overview
- Problem
Our client had already built a strong foundation with their core lending platform, but one critical piece was holding them back: the credit scoring module. At first glance, the system worked — it could evaluate borrowers and help decide whether to issue loans.
But beneath the surface, there was a problem.
Each new financial institution had its own scoring criteria. Adjusting the solution to them required developers to make changes directly in the code. That meant they were constantly pulled into repetitive tasks, tweaking rules for age, salary, loan history, or expenses, instead of focusing on innovation.
On top of that, monolithic architecture didn’t allow to introduce changes on the go.
This rigidity created a bottleneck. For our client’s customers, it meant slower onboarding and limited flexibility. For our client themselves, it meant higher costs, wasted resources, and a product that couldn’t scale as quickly as the market demanded.
- Our solution
We added the new logic to conveniently tweak credit scoring rules directly from the tool’s interface. This allowed our client’s staff to adapt them to each new customer without involving developers.
We also extended the app’s functionality to process smaller loans automatically. It collected various data on applicants from their profiles and other sources and applied a grading algorithm. For example:
1. Employment stability:
– Full‑time job with 3+ years at current employer → +15 points
– Frequent job changes in the past 2 years → –10 points
2. Savings & assets:
– Maintains a savings account with consistent deposits → +10 points
– No savings or assets → –10 points
3. Payment behavior:
– No late payments in the past 12 months → +15 points
– Multiple late payments in the past year → –20 points
4. Credit utilization:
– Uses less than 30% of available credit → +10 points
– Maxed out credit cards or overdrafts → –15 points
(the reference doesn’t illustrate the actual scoring criteria and algorithm)
Summarizing results, the tool worked out the final grade and made a decision on approving/rejecting the loan. Riskier or more complex loans still required manual review after the automatic assessment.
Pursuing our vision of tech efficiency, we also re‑engineered the platform’s architecture. We broke the monolithic application into microservices that made the system easier to support, customize, and connect with other tools.
- Workflow
The development team operated fully remotely, following Agile methodology with structured sprints and bi-weekly demos to ensure continuous feedback and alignment. Transparent communication and progress tracking were maintained through Jira for task management and Slack for daily collaboration.
What our client achieved
67%
faster solution rollout for new partners
75%
faster application processing
99%
Accurate borrowers’ evaluation
What our client achieved
67%
faster solution rollout for new partners
75%
faster application processing
99%
Accurate borrowers’ evaluation
Repeat the success
Want the same results?
Leave us your details and we’ll contact you soon to understand your business needs
About you
About your project
Related projects
1
2
3
4
5
6
7
8
9
10
11
12
13
Data architecture
1
2
3
4
5
6
7
8
9
10
11
12
13
Data modeling
Data warehouse development
1
2
3
4
5
6
7
8
9
10
11
12
13
Custom software development
1
2
3
4
5
6
7
8
9
10
11
12
13
Retail & eCommerce
Dedicated team
Related projects
1
2
3
4
5
6
7
8
9
10
11
12
13
Data architecture
1
2
3
4
5
6
7
8
9
10
11
12
13
Data modeling
Data warehouse development
1
2
3
4
5
6
7
8
9
10
11
12
13
Custom software development
1
2
3
4
5
6
7
8
9
10
11
12
13
Retail & eCommerce
Dedicated team