As 9 out of 10 loan applications are binned across the formal financing industry, startups such as BankSathi, GoodScore and Credgenics are betting on AI-led advisory and services to help reduce rejections.
This is not the first time startups are trying to solve this problem. Credgenics, which recently launched its product Fixmyscore, had earlier attempted to build a similar offering in 2021 but shelved it due to cost issues. Chennai-based CreditMantri and Paisabazaar have been running credit score improvement products for years.
But now, with AI, the new generation of startups is reimagining the product flow, while older players are streamlining their processes.
“AI can get 70 to 80% of the backend work done today; earlier, one needed to deploy agents who would work with borrowers in a step-by-step process, but the lifetime value of the customer did not justify the customer acquisition cost,” said Rishabh Goel, cofounder, Credgenics, a debt collection startup that has recently ventured into this space with Fixmyscore.
The right time
Additionally, multiple macroeconomic factors have also come together to make this business relevant again, according to industry insiders.
Banks have tightened their underwriting processes, defaults in unsecured lending have gone up in recent times, and technology innovation through AI has become mainstream. Founders believe this is the right time to build scalable, tech-led platforms to address this issue.
Jitendra Dhaka, cofounder and chief executive officer of loan sourcing platform BankSathi, said there are around 40 to 45 million borrowers whose loan applications get rejected every month, highlighting the size of the opportunity.
“There is a huge demand, especially from consumers in smaller cities and towns who are not aware of how credit scoring works. Our platform helps explain the process to them and offers ways to improve their scores over a three- to six-month period,” Dhaka said.
Bengaluru-based Banksathi has raised $2.3 million in funding from Kotak Securities, Kunal Shah and others.
Banksathi is doing this with almost no manual intervention on the backend, he added. Around 10,000 users are using this service every month.
AI at scale
For Paisabazaar, AI is proving useful as it processes data at a scale impossible for humans to manage. Over the last 10 years that Paisabazaar has been offering credit awareness as a service, more than 57 million people have accessed it.
Mukesh Sharma, chief technology officer of Paisabazaar, told ET that the company is using AI to build a detailed understanding of a borrower’s credit history, creating fully AI-generated videos to explain the challenges to customers, and offering an AI-led chat service to assist them.
“120,000 people generated these video score reports with us last year, and 6,000 people engage with the chat feature every month,” Sharma said.
Stressing further on the role AI is playing, Sharma said there is significant upfront investment going into AI-led processing capabilities, but it is helping them process data for customers in real time, making the entire process smoother.
“Previously, we used to run the models in batches every week; now all this is happening in real time,” he said.
Manual intervention still needed
However, Goel of Credgenics pointed out that there is still a need for manual intervention if fintechs want to help borrowers resolve defaults.
With AI, customers can be shown the loans they have not repaid and be told what can be done to improve their scores, but resolution still needs to happen with banks or lenders, where processes remain manual.
“Currently, 80% of the work is being done by AI; the remaining 20% will be digitised only when banks create standard processes around resolutions. At present, that leg is fully manual and decided on a case-by-case basis,” Goel said.
Dhaka pointed out that Banksathi is building solutions where AI-generated emails can be sent to banks for resolution. However, he noted that communication between the customer and the bank typically happens outside the Banksathi ecosystem, a space where the platform has limited control.
Changing the cost game
With customers already coming to these platforms for their credit needs, founders believe they can keep customer acquisition costs low and use this score improvement product as a cross-sell.
Additionally, as this is being offered as a subscription product, it adds directly to revenue.
An investor who has evaluated the sector pointed out that if a customer improves their credit score, it is ultimately the fintech that benefits by converting the customer into a borrower.
Credgenics, for instance, has built integrations with around 250 to 300 lenders to help them settle with defaulting borrowers. The company has also built a step-up credit card to help new-to-credit customers build their scores in a secured manner.
