Carrington Labs launches Cashflow Score on Snowflake

Carrington Labs has launched its Cashflow Score as a Native App on the Snowflake Marketplace, giving Snowflake users in Australia access to the lender's cashflow underwriting tool within their existing data environment.

The product is aimed at lenders that already store borrower transaction data in Snowflake and want to apply cash flow-based credit assessment without moving data outside the system. It removes the need for external data sharing or a bespoke model build.

Cash flow underwriting has attracted growing attention from banks and non-bank lenders seeking ways to assess applicants with limited or incomplete traditional credit files. Transaction-level bank data can offer a different view of a borrower's income, spending patterns, and financial resilience, particularly for thin-file and no-file applicants.

The app gives users access to Carrington Labs' Cashflow Score 2.0 model. The updated version adds categorisation of bank transaction data and clearer score drivers alongside the score itself.

Scores are returned on a scale of 1 to 100, with 100 indicating the highest credit quality. Lenders can use the output in their underwriting workflows already running in Snowflake, rather than exporting information to another system for analysis.

Inside Snowflake

The tie-up reflects a broader push by data and analytics suppliers to place tools directly inside customers' cloud data platforms. For lenders, this can reduce the operational workload of deployment and address concerns about handling sensitive financial information across multiple environments.

Many lenders see value in bank transaction data but struggle to put it into production in a way that delivers measurable results. Embedding the score as a Native App is intended to simplify that process for institutions already using Snowflake.

Carrington Labs also presented the move as a practical option for lenders seeking to update underwriting processes without lengthy integration work. Rather than building custom cash flow models from scratch, they can install the application from the marketplace and connect it to existing workflows.

The approach may be particularly relevant in consumer and small-business lending, where access to richer, more current financial data has become increasingly important as lenders refine risk selection. Traditional credit bureau data remains central to many decisions, but alternative data sources have become more prominent in efforts to broaden access to credit while managing losses.

The app is available to Snowflake users in Australia, with other regions expected to follow. Carrington Labs did not specify which markets would come next.

Model update

The Snowflake app launch coincides with the rollout of Cashflow Score 2.0. The addition of transaction categorisation and more explicit score drivers points to a focus on explainability, which remains important for lenders using automated decision tools in regulated credit settings.

Explainability has become a significant issue in credit analytics as banks and non-bank lenders face pressure to understand and document how model outputs affect decisions. Tools that show what is influencing a score can be easier to integrate into governance and oversight processes than black-box systems.

Snowflake has increasingly positioned its marketplace and native application framework as a way for software and data providers to distribute products into customer environments without requiring large-scale data transfers. Financial services is one of the sectors where that model can appeal because of restrictions and sensitivities around customer information.

For Carrington Labs, the Snowflake listing provides another channel to reach lenders already investing in cloud-based data infrastructure. It also places the company's underwriting product closer to the systems that many lending teams now use to manage and analyse data.

Kasey Kaplan, Deputy Chief Executive Officer of Carrington Labs, outlined the company's rationale for the launch in a statement accompanying the announcement.

"Lenders don't need another abstract analytics project. They need practical ways to improve credit risk performance inside the systems and controls they already use," said Kaplan. "Making Cashflow Score available on Snowflake Marketplace gives clients a faster path to cash flow underwriting, while helping them keep data within their own environment and maintain control over how the score is used in their lending workflow."