Funnel & conversion

Why do personal loan ad campaigns lose money, and how do I fix it?

Most personal loan campaigns lose money because they optimise to form submissions, so platforms buy cheap unqualified leads. Other leaks are one landing page for every product, loose keyword match with few negatives and paid bidding on queries organic already owns. Fix it by optimising to approvals and splitting pages and queries by intent.

  • Updated 19 November 2026
  • By Ayan Chakraborty

No cost, no obligation - a written audit within 5 working days.

By Ayan Chakraborty, Founder, SBA Innovation Pvt Ltd, Kolkata. Last updated . Reviewed by Ayan Chakraborty.

Why do personal loan ad campaigns lose money, and how do I fix it?

Personal loan keywords are among the most contested in the category, so the auction punishes sloppy structure. The accounts we inherit are rarely badly managed; they are optimised for the wrong endpoint. When bidding learns from form submits, it buys the cheapest submits, which are often the least likely to be approved, and cost per approved application rises even as cost per lead looks healthy.

The fix is structural. Send approved applications back to the ad platforms so bidding learns from outcomes, build one landing page per product, intent cluster and language, run exact and phrase discipline with weekly search-term pruning, and clear every asset through compliance before submission so you stop rewriting after rejections. Finally, let organic take the queries paid should stop paying for. Judge the whole programme on blended cost per approved application.

Where personal loan campaigns leak, and the fix
LeakSymptomFix
Optimised to form submitsCheap leads, low approvalsImport approvals as conversions
One page for every productWeak relevance, low completionOne page per product, intent and language
Broad match, few negativesWasted spend on irrelevant queriesWeekly search-term pruning
Copy rewritten after rejectionsDelays and disapprovalsCompliance review before submission
Paid and SEO overlapPaying for queries organic ownsMove those queries out of paid

From our client work: when paid stopped bidding against organic

For a personal loan NBFC, 91% of applications came from Google and Meta paid at the start, and blended cost per approved application had risen from ₹1,410 to ₹2,060 over four quarters. Over ten months organic came to hold 61 queries in the top three, paid was rebuilt to stop bidding on them, and blended cost fell to ₹1,421 (−31%) with media spend roughly flat; the SEO and funnel work contributed alongside it.

Method: Engagement 019: blended cost is (media spend + our fee) ÷ all credit approvals from every channel; a paid-only cost over the same window moved from ₹2,060 to ₹1,884.

What did not work: Two of three long-form comparison hubs never ranked against aggregators; we stopped that cluster in month six.

Read case study 019 Personal loan NBFC, baseline and assumptions.

Primary sources

FAQ

Frequently asked questions

Is my cost per lead too high?

Judge cost per lead only alongside cost per approved application. A higher cost per lead with a much higher approval rate can be cheaper per approval, so optimise to the approval figure.

Should I stop broad match?

Use it only with strong signals and tight negatives. Many lender accounts do better with exact and phrase match, weekly search-term review and offline conversions guiding bidding.

How much of this is a landing page problem?

Often a large share. If one page serves every product and keyword theme, relevance and completion suffer. Build a page per product and intent, and test form length on completed applications.

Where does organic fit?

Organic takes queries where you rank well, so paid can stop paying for them and focus on gaps. In our personal loan engagement this was a main driver of lower blended cost.

Next step

Want this checked against your own campaigns?

The free Growth Audit reviews your funnel, tracking and compliance gaps and returns a written plan within 5 working days.

  • Free of charge
  • Delivered in 5 working days
  • No lock-in