Score the next payment on every Avant loan and card account.

Pave Cashflow Scores predict near-term risk, meaning the next few installments, the next statement and the next pull, from the bank data your applicants already link. Validate the lift on Avant’s own borrowers in 2–3 weeks, before you change a line of policy.

Scope a backtest
1B+
monthly transaction and outcome events on the Pave network
10,000+
cashflow attributes, from unified bank, credit and performance data
2–3 weeks
from data to a measured read on your own book
0
changes to your credit policy to get started

Four places cashflow could earn a spot in Avant’s decisions

Two scaled books, one customer.
Avant serves the same middle-income customer with a personal loan and a card. Who gets the second product, and at what line, is a near-term decision about someone whose cashflow you can already see.
A new card program brings a new population.
The NASCAR co-branded card launches in January 2027. Those applicants arrive through a different door than Avant’s existing channels, and the first line assignments get made before that population has any performance history.
The richest signal raises the hardest questions.
Transaction-level data carries the most repayment signal, and the most fair-lending scrutiny. Every attribute that goes near a decision has to be explainable, and some categories shouldn’t be used at all.
A strong model sets a high bar.
Avant’s underwriting just supported its first AAA-rated personal loan securitization. Any new signal has to show incremental lift against that, on the same borrowers, before it earns a place.

How this fits what Avant already runs

Scope a backtest
Start outside the approval decision.

Behavior scores for portfolio monitoring and loss forecasting let a new signal prove itself without changing who gets approved. Approval use comes later, if the lift is there.

Explainable attributes, your selection.

Every one of 10,000+ attributes is readable. Your team chooses which go into a model and which categories stay out.

No new step for the applicant.

Pave scores the same permissioned bank data your applicants already link. Nothing changes in the application flow.

One score per decision.

Installment risk, credit line assignment and payment timing each run on a score built for that risk event, rather than one number reused across all three.

THE TEST

How we’d prove it on Avant’s book

We won’t ask you to take a published result on faith. We scope the risk event with your team in one call, say 60+ DPD on personal loans or early delinquency on new card accounts. You send de-identified bank data with the decision and outcome on each application, no PII. We score it on a holdout you control, against the decision you actually made, and show the incremental lift overall and inside your existing risk tiers.

2–3 weeks
to a measured read on incremental lift, across both the personal loan and card books
0
changes to credit policy or the application flow to run the test

Pave for Avant’s loan and card books

Personal Loan Score
Predicts whether the next few installments clear, separating borrowers your current tiers treat as alike. The risk event and window are set with your team.
Cashflow Attributes
10,000+ stability, affordability and repayment signals, each one explainable, available as model inputs rather than a black-box score.
Income Prediction
Predicts the next deposit and when it lands, so affordability is measured against money that actually arrives.
API or Snowflake
Scores and attributes over API into your decision platform, or delivered to the Snowflake warehouse your analysts already use.
Credit Card Score
Reads current cashflow for initial line, line increase and early-deterioration decisions, including on a new card population.
Ongoing Monitoring
Scores refresh as cashflow moves, for portfolio monitoring and loss forecasting across both books. The same score family covers small dollar, cash advance, SMB charge card and SMB ACH risk.

The Pave Impact

Proven before you commit: we validate the lift on your own book in 2–3 weeks, before you change a line of policy.

Scope a backtest
Performance compounds with each risk event on the Pave network
Over 1 billion monthly transaction and outcome events feed back into our models, sharpening categorization, attributes and scores.
Customized for your book
We build a custom score around your borrowers and the specific risk event you're underwriting—no black box, no one-size-fits-all.
We become an extension of your risk team
We watch the lift, spot drift early, tune thresholds, and ship new versions—always finding ways to improve performance before you have to ask.
‍

See the lift on Avant’s borrowers before you change anything.

Send historical personal loan and card applications with outcomes, and get a read against your current decisions in 2–3 weeks. No policy change, no new step for applicants.

Scope a backtest

Scoped in one call

Measured against your current decision

No policy change to start