1. Auto approve / continue
Passes eligibility, affordability and risk rules, then continues to contract/disbursement checks.
LendGo platform
Decision engine demo
Shows the risk matrix logic behind each decision, with bank-statement transaction signals used to assess affordability, behaviour and repayment pressure.
2. Auto decline
Hard stops such as no consent, underage, confirmed tamper or strong compliance hit stop the journey.
3. Edge-case review
Borderline, missing or policy-sensitive cases are the exception path while thresholds are finalised.
Current result
Auto approve / continue
Account AUTO-APPROVE is processed through transparent rules. Every stop, continue or review decision has visible reason codes for audit and customer-service follow-up.
Affordability
R 8 500
Existing instalments
R 2 500
Arrears
R 0
Risk matrix logic
The matrix shows how each signal is evaluated, where the evidence comes from, and how it affects auto approve, auto decline or edge-case review.
Eligibility
Age and consent
Consent captured and age gate passed.
Evidence source
Application intake
Auto-decline hard stops before paid checks.
Bank statement affordability
Verified income and disposable cash
R 8 500 surplus; R 7 500 after requested loan.
Evidence source
Bank statement income and affordability extraction
Auto-approve only when cash buffer remains healthy.
Bank statement commitments
Recurring instalments vs income
14% of verified income is already committed.
Evidence source
Categorised debit orders, instalments and recurring payments
Route pressure cases to decline/review before disbursement.
Bank statement behaviour
Arrears and failed-payment pressure
R 0 arrears across 0 account(s).
Evidence source
Arrears balances, failed collections and account behaviour
Measure willingness-to-pay risk from actual account behaviour.
Bank statement behaviour
High-risk discretionary spend
0% of monthly income is high-risk discretionary spend.
Evidence source
Categorised gambling/lottery and similar transactions
Identify behavioural risk that a bureau score may miss.
Bank statement behaviour
Transfer and cash-out pressure
0% of income moved through transfers/outflows.
Evidence source
Large transfers, cash movement and unexplained outflows
Flag hidden obligations or cash-flow leakage for review.
Bureau and obligations
Debt load, active accounts and bureau score
65% debt-to-income; XDS score 640.
Evidence source
Credit bureau plus supplied account records
Combine bureau history with live cash-flow evidence.
Fraud and compliance
Statement integrity, AVS and AML/PEP
No confirmed hard-stop fraud/compliance result in this sample.
Evidence source
Vendor checks and compliance responses
Auto-decline confirmed hard stops; review unresolved ownership signals.
Bank statement transaction analysis
Bank statements help verify live cash flow, not just historic bureau behaviour.
1. Income
Verify salary/income deposits and consistency.
2. Commitments
Detect debit orders, instalments and recurring obligations.
3. Cash buffer
Check disposable cash after expenses and requested loan.
4. Behaviour
Flag arrears, failed-payment pressure, gambling and large outflows.
5. Decision
Convert transaction signals into reason codes and routing.
Feature ratios
Flags are visible and explainable.
Loan-to-affordability
Review line: 0,293
Instalment-to-income
Review line: 2,308
Debt-to-income
Review line: 5,472
Arrears-to-income
Review line: 3,058
Business value
Decision proof that can become delivery logic
The demo shows how each application can be routed immediately, while keeping the evidence and reason codes needed for audit, customer support and later model improvement.
Version
Proof
Sample rows
5
Hard-stop rules
0No flags in this group.
High-risk reasons
0No flags in this group.
Edge-case reasons
0No flags in this group.
Validation limits
Proof result, not a final production ML claim.
AUC
0.875
Gini
0.750
KS
0.750
Rows
5
These metrics prove the mechanics only. Production validation needs a larger labelled repayment dataset with paid, paid late, failed collection, defaulted and settled outcomes.
Next data needed
Needed before reliable trained-model accuracy.
Paid on time
Paid late
Failed collection
Defaulted or written off