By Ayan Chakraborty, Founder, SBA Innovation Pvt Ltd, Kolkata. Last updated . Reviewed by Ayan Chakraborty.
How do I calculate CAC per disbursal for an NBFC?
Divide your total acquisition cost for a period by the number of loans disbursed in that period. Total cost normally includes paid media, agency and tool fees, and sourcing payouts such as DSA commissions. Add telecalling and sales cost if you want a fully loaded figure, and say clearly which one you are quoting, because the two can differ widely.
Disbursals lag applications, so a loan approved this month may disburse next month. For a steady business a monthly figure is usable; for a fast-growing or seasonal one, match cost to the cohort of applications that produced the disbursals. Keep the definition fixed across periods. Compare CAC per disbursal with the lifetime margin on a loan to see whether acquisition pays back, and report it beside cost per approved application to see where the funnel leaks.
| Line | Value | Note |
|---|---|---|
| Media spend | ₹3,00,000 | Google and Meta invoices excluding GST |
| Agency and tool fees | ₹50,000 | Fees for running acquisition |
| Telecalling and sales cost | ₹1,20,000 | Include for a fully loaded number |
| Disbursed loans | 49 | From the LOS, same period |
| CAC, media and fees only | ₹7,143 | ₹3,50,000 ÷ 49 |
| CAC, fully loaded | ₹9,592 | ₹4,70,000 ÷ 49 |
What is the difference between cost per approved application and CAC per disbursal?
Cost per approved application stops at the credit decision; CAC per disbursal continues to money leaving the lender. The gap between them is the disbursal rate, the share of approved applicants who actually take the loan. A lender with a low disbursal rate can have a healthy cost per approval and a poor CAC.
That is why we report both. If cost per approval is fine and CAC per disbursal is high, the leak is after approval: slow documentation, offers that do not match what was advertised, or callers who lose borrowers between sanction and payout. If cost per approval is high, the leak is earlier: traffic quality, landing page or form. Use the calculator to model each stage, then fix the stage with the biggest drop.
From our client work: counting approvals versus disbursals
In the personal loan NBFC engagement we counted approvals at credit approval in the LOS and reported disbursals separately, because disbursal counts ran roughly 8 to 11% lower. Showing both stopped arguments about which figure was right.
Method: Engagement 019: approvals deduplicated by PAN and excluding re-applications within 30 days; disbursals reported in the monthly dashboard.
What did not work: A blended figure alone hides how much of the fall came from organic approvals, which carry no media cost; we report a paid-only figure beside it (₹2,060 to ₹1,884 over the same window).
Read case study 019 Definitions and footnotes.