Collection Efficiency for Lenders: Formulas and Levers
Collection Efficiency for Lenders: The Metrics, Benchmarks and Levers That Actually Move It
Last reviewed: November 2026
Collection efficiency is the share of the amount due in a period that a lender actually collects, usually calculated as collections divided by billed demand. It is the headline collections number in most board packs and investor presentations, yet two lenders can report the same figure while running very different books.
This guide is for Heads of Collections, CFOs and CEOs. It gives you the formulas with worked arithmetic, explains why published benchmarks rarely compare like with like, and sets out the levers that genuinely move the number.
Key Takeaways
- There is no single collection efficiency formula, so the same month can honestly be reported as anything from about 86% to 105%.
- Current-demand efficiency is the cleanest health signal, because it excludes arrears and prepayments.
- Roll rates and cure rates by bucket show deterioration before headline efficiency drops.
- From 1 January 2027, RBI recovery conduct rules narrow aggressive tactics, so gains must come from targeting and early intervention.
How Is Collection Efficiency Calculated?
Every version divides money collected by money due, but each includes different items. Choose one definition for management reporting and keep it fixed.
| Metric | Formula | What It Tells You |
|---|---|---|
| Current collection efficiency | Collections against current demand / current demand | Share of this month's EMIs paid on time |
| Collection efficiency including arrears | (Current collections + arrears collected) / current demand | Whether old overdues are being cleared |
| Headline collection efficiency | (Current + arrears + prepayments) / current demand | The number often seen in presentations |
| Total-dues efficiency | (Current + arrears collected) / (current demand + opening overdue) | How much of everything owed was recovered |
| Roll-forward rate | Accounts moving to a worse bucket / opening accounts in bucket | Where the book is deteriorating |
| Cure rate | Accounts returning to current / opening accounts in bucket | How well early collections work |
| Cost to collect | Collections cost / amount collected | Efficiency of the collections spend |
The first metric is the cleanest test of borrower behaviour. The others are useful, but they can rise even while new delinquency grows.
How Different Can the Answers Be? A Worked Example
This is an illustrative example with assumed numbers. An NBFC bills ₹50 crore of EMIs in a month and starts the month with ₹8 crore already overdue.
During the month it collects ₹46.5 crore against current EMIs, ₹3.2 crore of arrears and ₹2.8 crore of prepayments and foreclosures. Here is the same month under four definitions.
| Definition | Calculation | Result |
|---|---|---|
| Current efficiency | ₹46.5 crore / ₹50 crore | 93.0% |
| Including arrears | (₹46.5 + ₹3.2) / ₹50 = ₹49.7 / ₹50 | 99.4% |
| Headline, including prepayments | (₹46.5 + ₹3.2 + ₹2.8) / ₹50 = ₹52.5 / ₹50 | 105.0% |
| Total-dues efficiency | ₹49.7 / (₹50 + ₹8) = ₹49.7 / ₹58 | 85.7% |
All four numbers are arithmetically correct. Only the 93.0% tells the board that 7% of this month's billing, ₹3.5 crore, went unpaid.
Now look at bucket movement. Suppose 10,000 accounts sat in the 1 to 30 days past due bucket at the start of the month.
- 7,200 paid up and returned to current: cure rate = 7,200 / 10,000 = 72%.
- 1,900 stayed in the same bucket: stabilisation rate = 19%.
- 900 moved to 31 to 60 days: roll-forward rate = 9%.
The three rates add to 100%. If the roll-forward rate climbs from 9% to 12% while headline efficiency stays above 100%, the warning is in the roll rate.
Finally, cost. If the same month's collections operation costs ₹1.2 crore and recovers ₹49.7 crore of current dues and arrears, cost to collect is ₹1.2 / ₹49.7 = 2.41%.
Split that by channel and the picture sharpens. Suppose field teams spend ₹80 lakh to recover ₹6 crore, a cost of 13.3%, while digital and calling channels spend ₹40 lakh to recover ₹43.7 crore, a cost of 0.92%.
How Should You Segment the Number?
A single portfolio figure hides more than it shows. Segment collection efficiency in the same way you segment credit risk.
- By product: secured and unsecured books behave differently, so blend them only at the top.
- By origination month: newer cohorts with weak early payment point to underwriting or onboarding problems.
- By sourcing channel: a DSA or partner whose loans miss first EMIs needs attention before volumes grow.
- By geography and branch: local events, such as floods or crop failure, show up here first.
- By payment mode: autopay, payment link and cash collections have very different failure patterns.
Ownership matters too. Finance and collections should agree one definition, reconciled to the loan ledger, so the number in the board pack matches the books.
What Benchmarks Should You Actually Use?
Use your own history on a fixed definition as the primary benchmark. Public figures help only when you know exactly how they were calculated.
Listed lenders often publish collection efficiency, but definitions vary on prepayments, arrears and restructured accounts. Product mix also matters, since secured, unsecured, microfinance and MSME books behave very differently.
If you need reference points for target setting, build them internally. The ranges below are illustrative only, for a hypothetical lender, and are not industry benchmarks.
| Metric | Illustrative Internal Target | How to Set Your Own |
|---|---|---|
| Current efficiency | 94% to 97% | Trailing 12-month median by product |
| 1 to 30 DPD cure rate | 65% to 75% | Best quartile of your own months |
| 1 to 30 DPD roll-forward | Below 10% | Cohort trend, adjusted for seasonality |
| Cost to collect | Set per product | Cost per rupee collected, by channel |
Banks and NBFCs already track RBI's special mention account buckets. For loans other than revolving facilities, SMA-0 covers overdues up to 30 days, SMA-1 covers 31 to 60 days and SMA-2 covers 61 to 90 days.
Does your board see the collection efficiency that matters, or the one that looks best?
Talk to Our Lending Technology Specialists
Visit Digiqt to see how we build collections workflows and AI agents for lenders.
Which Levers Actually Move the Number?
The strongest levers sit before the due date and in the first 30 days. Later-bucket effort recovers money, but it costs more and carries more conduct risk.
- Mandate quality at onboarding. A registered autopay mandate on the right account prevents many first-time misses.
- Due dates that match income. Aligning EMIs with salary or business cash-flow cycles reduces technical bounces.
- Pre-due reminders. A reminder a few days before the due date costs little and catches insufficient balances.
- Frictionless payment. A UPI link in a reminder converts intent into payment immediately.
- Prioritised early follow-up. Scoring accounts by risk and likely response focuses effort where it changes outcomes.
- Hardship pathways. Early restructuring or rescheduling within policy can cure accounts that pressure never would.
Prioritisation is where data pays off. Digiqt's collections prioritization AI agent ranks accounts by risk and response, while an early delinquency warning AI agent flags stress before the first missed EMI.
For high-volume reminders, voice agents for personal loans can run consistent, recorded conversations within permitted hours.
What Should the Weekly Collections Review Cover?
A weekly review keeps levers tied to results. Keep it short and use the same pack every week.
- Current efficiency month to date, against the same point last month.
- Cure and roll-forward rates for the 1 to 30 and 31 to 60 day buckets.
- Bounce reasons for failed autopay debits, such as insufficient funds or mandate errors.
- Promises to pay made, kept and broken, by channel.
- Cost to collect by channel, and complaints per 1,000 contacts.
Each item should have an owner and one agreed action. Without that, the review becomes a reporting ritual instead of a management tool.
Which Reporting Mistakes Mislead Leadership?
These reporting habits make a deteriorating book look healthy.
- Counting prepayments in the headline. Foreclosures inflate efficiency while hiding current misses.
- Shrinking the denominator. Excluding restructured or written-off accounts without disclosure flatters the ratio.
- Reporting only the month-end figure. Collections bunched into the last days can mask slow early-month payment.
- Paying agents only on amount collected. RBI's 2027 rules bar incentives that induce harsh recovery practices.
What Is Changing Next?
The RBI's recovery conduct directions take effect on 1 January 2027, covering contact hours, call recording, agent certification and incentives. Collections strategies that rely on volume of contact will need redesigning.
Credit information reporting also moved to a weekly cycle from 1 July 2026, with reference dates on the 9th, 16th, 23rd and last day of each month. Missed payments now reach bureau records faster, which raises the value of curing accounts early.
Want collections effort focused on the accounts where it changes the outcome?
Talk to Our Lending Technology Specialists
Visit Digiqt to see how we build lending platforms, dashboards and collections AI agents.
