Tool · Analytics
Decide with the data your lending already produces.
From the first application to the last payment: where applicants drop off, why loans are declined, how your loan book is performing and how healthy your finances are. One place for your commercial, risk and finance teams.
Why it matters
The richest data goes unused.
Onboarding and credit analysis record everything, and nobody looks at it once the decision is made.
Finance works from spreadsheets.
The loan book and the numbers that matter to the board are rebuilt by hand every month.
Decisions don't learn.
Nobody connects how a loan was approved with how it was paid.
What it does
The application funnel
Applications started and completed, drop-off and time spent at each step, by channel and product.
Biggest dropDocuments, by channel: App 31% · Website 44% · WhatsApp 36%
Reasons for declineLast 90 days
Debt-to-income above limit34%
Scorecard below cutoff28%
Incomplete documents17%
Credit bureau history14%
Other7%
By product
ProductApprovedDecisionOverrides
Personal loan64%2h 50m3.1%
Vehicle loan58%5h 10m6.4%
Microcredit71%12m1.8%
Paying on time, by score band and month of approvalPersonal loan · all channels
LowerHigher
Worth a lookBand C approved in April: 79% on time, the lowest of the half.
Late-payment rate · 12 months10.7% now
Share of the loan book with any payment late, month by month.
How loans move between late groupsLast month
Current → 1–30 days2.1%
1–30 → 31–60 days18%
31–60 → 61–90 days34%
61–90 → 90+ days46%
Each month's loans · late at month 6By month of approval
Where the book is concentratedBy product
Personal loans48%
Vehicle loans31%
Microcredit21%
Provisions$1.32B · 7.5% of the book
Disbursements against collectionsDisbursedCollected
Yield by productAnnual
ProductYieldRiskNet
Personal loan31.2%2.4%28.8%
Vehicle loan22.8%1.9%20.9%
Microcredit44.1%4.6%39.5%
Payments expectedNext 4 weeks · $1.21B
Commercial, risk and finance each filter by product, channel, team and date.
Take any view into your board pack or regulatory report.
Your own tools read what the dashboards show.
How it works
- Back to the start
AI in this tool
Patterns pointed out. AI highlights what deserves attention, such as a step where many good applicants give up, a rule that declines customers who would have paid, or a group of loans starting to slip.
Better rules, suggested. Those findings become suggested changes in the decision engine or collections, ready for your team to review and approve.
Questions in plain words. Through MCP, your AI assistant can ask the analytics directly, for example "What was our late-payment rate by product this month?", within your team's permissions.
Use it with Cowally, or on its own.
Use it from the back office, connected to every other part.
Your developers plug it into your own product.
See API By MCPYour people and AI agents reach it, with roles and rules.
See MCPQuestions about analytics
Everything your onboarding, decision engine, loan book, payments, collections and accounting record.
Yes. Loan book health and financial health are built for finance and risk, alongside the commercial views.
No. The dashboards are ready to use; your data team can go further through exports and the API.
Yes, through exports and the API.
See what your lending is telling you
In one session we'll show the funnel, the decisions, the loan book and the financial views with data like yours.