See what your applicant actually has left
Pouch turns the categorised transactions you already pull at application into a true affordability picture: real income, itemised commitments, genuine discretionary behaviour with internal transfers stripped out, and the applicant's forward free-cash position on every date in the repayment schedule.
Affordability · resolved to the day
Applicant 2087 · £4,800 over 6 months
Clears every due date · instant at application
01
Affordability built on guesswork
Today you assess consumer affordability off gross income and a bureau score.
What your affordability check reads
Income · stated
Bureau score · the past
Left each month after real commitments
· never asked ·
and you approve six instalments from this.
over-lend and eat the defaultunder-lend and lose the customer
02
How it works
How it works at application
Step 01
Push from inside your funnel
The applicant connects via open banking at application, exactly as they do today, and you push the categorised transactions to Pouch's endpoints.
Checkout · order £840pay in 6 · £140/mo
applicant connects via open banking · 9 seconds
SALARY · MONTHLY+£2,450.00
TRANSFER · OWN SAVINGS+£350.00
RENT · STANDING ORDER−£950.00
POST /v1/transactions · 202
returned while the checkout spinner turnszero added friction in the funnel
Step 02
The discretionary split
Pouch excludes internal transfers from spending, itemises mandatory commitments, and measures discretionary spend and savings instead of guessing them: the applicant's real monthly capacity to absorb a repayment, not their gross income.
The resolved position
the application said £2,800 · the detected salary cycle pays £2,450
rent · car finance · energy · detected cycles
Discretionary
£430
Savings
£190
both can flex toward a repaymenta committed bill cannot
measured from how they actually spendnot from what the form says
Step 03
Every due date, checked
The forward daily forecast returns the applicant's projected position on every due date in the schedule, resolved to the day.
Six due dates · position after each £140
Oct 12
+£212
Nov 12
+£186
Dec 12
+£238
Jan 12
+£94
tightest · still clears
Feb 12
+£205
Mar 12
+£241
the January squeeze, seen in Septemberabove water on every due date · known before you approve
to the day · not a multi-month band
Step 04
Fewer defaults, same approvals
The corrected picture works both ways: it catches the applicant your gross-income check would have approved into a default, and rescues the thin-file applicant it would have declined.
Their check said yes
gross £2,900 · score 681
Real capacity £86/mo · the £140 instalment fails by month 2.
Declined · a default avoided
Their check said no
thin file · score 588
Real capacity £620/mo · every due date clears.
Approved · a customer kept
fewer defaults · without cutting approvals
03
The demo
What you will see
04
FAQ
The questions your team will ask.
- We already use bureau data and open-banking affordability. What does this add?
- Those read the past. Pouch adds the forward daily free-cash position: whether the applicant's cash stays above water on every due date in the schedule, which no attribute set gives you. It sits beside your existing checks as an additional layer, not a rival score.
- How does this sit with responsible-lending expectations?
- It strengthens them. An itemised, auditable affordability picture, with internal transfers excluded from spending and commitments listed line by line, is exactly the evidence a responsible-lending review wants to see. Pouch is decision support beside your existing checks; the decision stays yours.
- How accurate is a daily forecast?
- The forward position is calculated from detected income and outflow cycles, not guessed, and the pilot exists so you can measure it yourself. We run the engine on your historical applicants and you compare the projection to what actually happened, on your own book, not our claim.
- What do we need to provide?
- Categorised transactions, which you already hold from the applicant's open-banking connection. A few months of history is enough to produce output, and a longer history sharpens the forecast. In the pilot, your lift is little more than a data export.
See it on your own data.
Book a demo, then judge it on your own historical applicants.