Fintech Plaid noted that the data that makes up a consumer’s complete financial picture is more complex than it has ever been before. According to a blog post from Plaid, income is now more variable, expenses tend to fluctuate, and financial activity is spread across a number of platforms. The Fintech firm questions why is so much of that data still missing from important credit decisions.
That’s why they’re introducing Plaid LendScore, a credit risk score that uses real-time cash flow data as well as account connection insights from the Plaid Network to give lenders an updated view of overall borrower risk.
With nearly 1 million connections made on a daily basis, Plaid’s network provides differentiated signals about a borrower’s financial health that can power “predictive models.”
Delivered via their consumer reporting agency, LendScore is described as being a smarter way to assess risk and offers consumers the chance to share a more “complete picture of their financial lives, helping expand access to credit.”
Traditional credit scores still matter, but they rely “on historical data and miss what’s happening right now in a borrower’s financial life.”
Lenders increasingly see them as just “one part of the picture.”
LendScore takes a different approach to evaluating credit risk by leveraging cash flow insights, “income patterns, and financial account connections to reveal a borrower’s real-time financial story.”
This makes LendScore suitable to use alongside “traditional scores to enhance credit decisioning.”
Their first model, LS1, is trained on a dataset of nearly “one billion transactions to predict the likelihood of default in 12 months for unsecured loans.”
Additionally, they partnered with FairPlay, a provider of AI fairness techniques used by banks and fintechs, “to complete an independent assessment of their model.”
After a borrower agrees to share their bank account data via Plaid Link, lenders can call our API to receive a “score from 1–99, along with adverse action reason codes to support FCRA and ECOA compliance.”
They built LendScore in partnership with unsecured lenders, which helped shape the attributes they had prioritized, the “structure of their reason codes, and how they gauged model performance.”
For those lenders, they have now generated millions of scores and “delivered a 25% lift in predictive performance compared to traditional credit data alone.”
Additionally, they claim to have driven a 10–20% relative risk reduction for subprime and near-prime borrowers without reducing originations. This could save personal lenders more than “$1 billion in credit losses and reduce the average APR for borrowers by 3 percentage points.”
LendScore isn’t just another model—it provides lenders the tools to assess creditworthiness with greater “precision, speed, and context.”
LendScore is said to be the only score in the market that incorporates network insights, which offers context “on the types of apps someone is connected to on the Plaid Network, within a certain time period, and how long they’re active on each service.”
With over 150 million people who have linked their accounts to 7,000+ apps and services, the model picks “up early indicators of changing financial habits and emerging risk.”
The Plaid Network sees nearly 1 million daily connections “across checking, savings, investment, and loan accounts.”
This gives their scoring model the breadth to capture a borrower’s complete financial life and the depth to “analyze key signals like balance trends, income stability, and recurring expenses.”
All of these data points are distilled into an actionable risk score, “helping lenders move from insight to decision faster.”
Used by 1 in 2 U.S. bank account holders, their familiar and secure account linking experience delivers conversion rates “up to 80% in lending flows—far exceeding industry averages.”
Built-in returning user experiences make it faster and safer “to connect accounts, reducing friction and boosting conversion even further.”
They believe in an open, connected financial system that drives innovation and expands “responsible” credit access.
Lenders now have access to a variety of alternative data scores in the market, which enables them to “choose one or multiple that best aligns with their credit strategy.”
Unsecured lenders can use Plaid‘s score to meet their full underwriting needs or alongside “existing credit scores to supplement risk processes.”
As open banking continues to scale in the United States, credit scoring should keep getting smarter.
Plaid concluded that they will continue evolving their model to reflect how borrower behavior changes.