HomeFinance NewsHere is how the fintech startups use artificial intelligence to ruffle lending market

Here is how the fintech startups use artificial intelligence to ruffle lending market

With fintech lenders pushing the boundaries of technology for potent risk management, significantly reducing the time taken to process a loan and removing the need for collateral, banks are sensing an increasing threat.

Profile imageBy Kubra Fatima  August 5, 2019, 2:26:42 PM IST (Updated)
Here is how the fintech startups use artificial intelligence to ruffle lending market
With an efficient credit model and smart use of technology, India's financial technology (fintech) startups are cashing in on customers outside the purview of traditional financial institutions. These fintech firms are using Artificial Intelligence (AI) to automate lending process and reducing risk by mapping previous payments behaviour of borrower.



According to the India FinTech Report 2019, there has been an almost three-fold growth in the number of fintech startups during 2015-2018. More than 1,300 new startups were added to the base number of over 730 already in practice.

Shaick Mohammed Rafi, a small trader from India's IT capital Bengaluru, is among the ones who have benefited by associating with fintech startups. Stiff competition from e-retailers meant his revenue was much less than what he had projected. Owing to this gap in his balance sheet, banks were denying him business loans.

He then learnt about the concept of SME (small and medium-sized enterprises) lending and registered on a Bengaluru-based fintech firm Innoviti’s platform. Based on data provided by him, the startup granted him a loan of about $3000 quickly. It was a big relief, he said.

With fintech lenders pushing the boundaries of technology for potent risk management, significantly reducing the time taken to process a loan and removing the need for collateral, banks are sensing an increasing threat. To tackle this, without swallowing fintech startups whole, banks are increasingly providing investments and looking at collaborations. While technology and speed are a startup’s strength, banks are strong with capital. The largest source of capital for fintech startups are banks, and banks face major competition from such lenders. So the system works in a way to eliminate threat.

Look at the numbers: Innoviti claims to have processed loans amounting to more than Rs 24,000 crore, and card EMI transactions to the tune of Rs1,500 crore in the last financial year. It also claims to have installed active terminals in more than 800 cities in India.

Some of the top banks investing in the space are State Bank of India, Yes Bank, HDFC Bank, Kotak Mahindra Bank and ICICI Bank. Manish Jain, a partner at KPMG and an industry expert, says, “Banks lock up startups to prevent them from going elsewhere, this is an advantage for them.”

Fintech platforms’ reach, backing from well-known investors and the confidence of large banks to lend to them have led to more and more Indians moving towards app-based financial transactions, a scenario previously unheard of in a largely cash-dependent economy.

According to a KPMG report, fintech transactions worth approximately $33 billion took place in 2016 in India. This amount is expected to touch the figure of $73 billion by 2020. Industry experts say fintech startups are causing two major impacts: lowering the cost of capital and easing access, which is leading to an acceleration in the debt market.

Operating principles

Lenders in the fintech space have a distinct edge over traditional financial institutions as they are removing time-consuming paperwork and background checks.

Innoviti uses AI to automate its lending process, by adding intelligence to data, substituting human action to make lending faster, more real-time and dynamic, and also reduce the cost of risk and access. Traditional borrowing is a time-consuming system of waiting in queues and submitting endless paperwork.

Rajeev Agrawal, Founder and CEO of Innoviti, says traditional lending business has a lot of friction at each step. “With fintech, you get aggregated access to small businesses and borrowers, and you can use intelligent data and automation to enable quick disbursal of loans."

There is a common thread agreed upon by lenders to study risk; using data and technology to analyse a person’s intent to pay the loan and their ability to pay it.

“Our lending strategy is based on, a) figuring out a situation where a buyer is about to buy something, and b) enabling the buyer to take a loan instantly when they’re short of cash,” he says. A part of Innoviti’s approach is to understand the timing of loan disbursal using a formula. “Can we, from his past payment data, predict his ability to pay in the future? And when we can predict his ability to make the payment, we can give him a loan.”

Innoviti claims to have about 30,000 small buyers on its platform. Their company philosophy moves away from the traditional lending formula that sees transactions only between two parties: a lender and a borrower. Instead, it introduced a “three-actor game”, which involves a lender, and a buyer and seller. Innoviti offers loans only when a buyer is about to purchase from a seller, without giving it directly to the buyer. By bringing the seller into the picture and mapping previous purchase behaviour, it controls the borrower’s actions, a point Agrawal says is “critical to lending”.

Backed by one of India’s top IT and ITES firm Infosys co-founder NR Narayana Murthy’s Catamaran, Innoviti in mid-2017 raised Rs120 crore in Series B funding from SBI-FMO, a Singapore-based equity fund and Bessemer Venture Partners, among others.


Non-performing loans (NPL) at this lending startup figures at less than 0.7 percent. They disburse small-ticket loans — just about Rs15,000-Rs20,000 — not high-velocity ones.

Tackling risk with technology

Fintech lenders are using technology to evaluate risk management by administering credit, predicting consumer behaviour and more.

Lending platform Faircent, for instance, has a systematic method in assessing, preventing and managing risk. Every registered borrower is identity checked, credit checked and risk assessed. This is done through a technology-driven process of verification that assesses a potential borrower on 120 criteria and 400 data points, including social, financial and personal data. The data is sourced from personal and financial documentation provided by the customer.

Rajat Gandhi, Founder and CEO of Faircent, a P2P startup, says, “Since traditional data points were either unavailable or not reliable for most of our users, we decided to leverage non-conventional metrics to assess the borrower creditworthiness.” Based on the borrower’s ability to repay a loan, their automated system suggests an appropriate loan amount and period.

Manav Jeet, MD and CEO at Rubique, a financial matchmaking platform, says they assess a borrower’s ability to repay a loan by crunching data points such as bank statements, past tax records, civil data and others.

Then, there is the intention. “There’s no technology that can help you gauge intention. How you do this is by using alternative data points, where someone is behaving in a particular fashion,” says Jeet. For example, tracking phone payments, mapping e-commerce transactions, social media accounts, and other similar data points.

Rubique uses technology for predictability, creating algorithms and using data science and AI to match lenders and borrowers. “With algorithms at one side and data at the other, we create magic, which is nothing more than a blend of conventional data points with alternate data points,” explains Jeet.

The Mumbai-based Rubique claims to have disbursed loans amounting to more than Rs3,500 crore to more than 2.2 lakh customers. It boasts of an approval ratio of more than 90 percent. Rubique is backed by big names such as Kalaari Capital and Emery Capital, who recently invested an undisclosed amount, estimated to be between $15-$20 million.

Fintech employs the simplest of innovations like chatbots to answer customer queries, uses retina and fingerprint scans for security, offers user-friendly interfaces, and automates a major portion of the process.

Importance of tech know-how 

According to Samir Kumar, Managing Director at Bangalore-based VC Inventus Capital, the tech component of a team, their knowledge of machine learning and their background in technology are key while considering investment in an organisation.

Kumar opines that an organisation’s formula and philosophy for risk management need to be airtight for investment to flow in. “We look at a team’s experience in managing risk. With technology involved, it is necessary for a risk officer to use machine learning to make better decisions,” he says.

Fintech lenders are also leading the way to steady financial inclusion as even low-income individuals and businesses are able to access credit. This is contributing to the central government’s financial inclusion plan.

Gandhi explains that about 600 million Indians lack access to credit while only 20 per cent has access to organised credit from legacy financial institutions; the rest depends on the unorganised sector, which can be exploitative. “Potential borrowers do not have reliable credit or financial history, which makes assessing their actual creditworthiness on traditional parameters almost impossible,” he said.

The fintech industry’s future is a captivating space to watch out for. “While we have most elements right, we are yet to figure out the ultimate hockey stick moment -- where the lego pieces fit together,” muses Agrawal.

Kubra Fatima is a freelance writer and a member of 101Reporters, a pan-India network of grassroots reporters.

Check out our in-depth Market Coverage, Business News & get real-time Stock Market Updates on CNBC-TV18. Also, Watch our channels CNBC-TV18, CNBC Awaaz and CNBC Bajar Live on-the-go!