Pensata logo

Behavioral Credit Infrastructure. Africa first.

Make finance see the person.

We measure the human character that credit files cannot — so that finance reaches everyone it should.

Pensata’s BYLAS™ Engine measures willingness to repay directly from behavior. We partner with lenders across emerging markets to turn rejected ‘thin-file’ applicants into a creditable, financeable population.

Request a MeetingRead the research

The Problem

A $5.7 trillion gap, built on missing files

The IFC estimates the unmet financing need of micro, small and medium enterprises in emerging markets at US$5.7 trillion. Hundreds of millions of applicants are declined every year — not because they will not repay, but because there is no file to read. Credit bureaus price history. They cannot price a person who does not have one.

The Insight

Credit has two axes. Data covers only one.

Capacity

Can they repay?

Increasingly commoditized. Bank transaction data, open banking, and platform records already measure cash flows well.

Willingness

Will they repay?

Historically inferred from repayment history or self-report. The BYLAS Engine quantifies it from revealed choices in a short gamified simulation.

How it Works

Three steps from behavior to underwriting

  1. 01

    Measure

    An applicant completes a short gamified simulation inside the partner’s existing application flow. Revealed choices, not questionnaires.

  2. 02

    Estimate

    A Bayesian engine extracts the applicant’s behavioural decision signals from the full record of choices. Together they form a quantitative signature of how this person makes financial decisions.

  3. 03

    Underwrite

    These signals power second-look underwriting on the partner’s declined applicants — pricing the willingness axis that transaction data cannot see.

Working With Lenders

We ask partners to believe nothing. We take the risk first.

Stage 1

Shadow validation

The simulation runs inside a partner lender’s live origination flow with no involvement in the credit decision. Predictions are matched against real repayment outcomes in a genuinely high-stakes context.

Stage 2

Second look

Once validated, we underwrite the partner’s rejected applicants on our own balance sheet, holding the full loan economics, with servicing and collections riding the partner’s existing processes. For the partner: incremental revenue from applicants they already decline, with zero change to their approved book.

We never ask a partner or an investor to believe what we have not risked our own capital on. Our own balance sheet is the argument.

Our balance sheet proves the model with our own money. Once proven, the underwriting infrastructure scales through partner lenders and funding vehicles, not through our balance sheet.

Partners own their raw data. No personal data is ever taken — scoring runs on anonymized keys via API.

Trust & Traction

Proven measurement. Published research. Protected IP.

  • Deployed measurement

    BYLAS measurement has been deployed with leading Japanese financial institutions through our Japan subsidiary, Beta Integral K.K., in executive and professional training programmes.

  • Published research

    Our methodology is documented in two SSRN papers: 'Beyond FICO' (2026), which sets out the BYLAS Score framework, and 'Loss Aversion Is a State, Not Just a Trait' (2026), a pre-registered field experiment with a Japanese financial-group training programme.

    Read the research

  • Patent-pending

    Patent applications filed in Japan and the United States, with rights consolidated in Pensata, Inc.

Measurement deployed with

  • MUFG Group
  • Daiwa Securities
  • Sumitomo Life

Backed by

Techstars