Fintech Plaid Bets on Credit Scores to Fuel Next Phase of Growth

Plaid is doubling down on its credit-scoring business with new AI models as the company aims to chart its next era of growth with a suite of products beyond its core business.

A year after the financial technology company released its first credit-risk tool, Plaid is launching a second version of LendScore with AI models that are meant to give lenders more insight about potential borrowers.

The product is part of Plaid’s efforts to stretch beyond its core business of connecting consumers’ bank accounts with financial apps such as Robinhood and Venmo. Now, the company is looking to spur its next phase of growth.

“For our phase two, we started thinking about what are the areas where we can really make the product quality a lot better,” Chief Executive Zach Perret said. “One big one for us was in credit score.”

Plaid rolled out its first credit-scoring tool, LendScore, last October. The model uses cashflow data—which includes information on how much someone has saved and how often they pay bills on time—to predict the likelihood that a borrower will pay back a loan.

LendScore 2, which is launching Tuesday, is designed to have a better rate of predicting someone’s likelihood of default on a loan payment. Lenders can use cash-flow data as a supplement or a replacement for traditional credit metrics such as the FICO score.

“Traditional credit data looks at your willingness to repay. Cashflow data looks at your ability to repay,” Plaid’s credit product lead said Michelle Young said. “These signals together give lenders a more holistic view to actually make a decision.”

The company is also debuting LendScore Arc, a credit risk model powered by AI technology. It learns from the order and timing of a borrower’s transactions to predict their ability to repay debt, and the likelihood of delinquency.

The new version of LendScore offers specialized models for auto and home lending that predict creditworthiness for people with those specific types of loans. It also has a short-term lending model designed for companies including buy now, pay later providers.

These specialized models are intended to attract new types of customers who might be apprehensive about new credit technology.

“It’s a very new thing,” Perret said. “We’re still talking to people about how it works. Lenders want to be very comfortable.”

So far, Plaid has seen the most traction among fintech lenders who are more apt to try new technologies, the company said. Consumer lenders are particularly interested because their loans are not collateralized and they want to know as much as they can about their prospective borrowers.

The tools are also supposed to help people who have little to no credit history get loans by providing cashflow data instead.

The underwriting business follows Plaid’s model of focusing on business-to-business relationships with a goal of improving consumers’ experience in the financial world, Perret said. Plaid’s core business is designed to make it easier for consumers to link their bank accounts with apps, and Perret has that goal in mind with some of the new products the company is currently developing.

Plaid has hinted about launching new products to help businesses and consumers with the payments process. It also has a growing anti-fraud business, and is doing more integrations with OpenAI’s ChatGPT and others to allow users to make payments through AI chatbots.

Plaid’s finance chief said earlier this year that Plaid is preparing to go public, and the company has stepped up product development in new areas including credit scoring and payments. Perret said he plans to announce several new products in the near future, some of which could come out this fall.

“We want to continue to push the edge of innovation,” Perret said. “My hope is that we’re partnering with people on most credit decisions in five years.”