Technology

Bucket-Wise Collection Strategy: Match Effort to DPD Risk

|Posted by Hitul Mistry / 16 Nov 26

Last reviewed: November 2026

Bucket-wise collection strategy is the practice of grouping overdue loan accounts by days past due (DPD) and giving each bucket its own channel, script, staff and escalation rules. Without it, lenders spend the same effort on a borrower who forgot one instalment and one who has stopped paying, and accounts quietly roll into costlier buckets.

This guide is for heads of collections and CROs at NBFCs, banks, housing finance companies and fintech lenders. It covers how DPD buckets map to RBI's special mention account (SMA) rules, why uniform strategies fail, a six-step workflow, a cost scenario and the metrics that show it is working.

Key Takeaways

  • Each DPD bucket needs a different channel, script and owner, because the reasons for non-payment change as accounts age.
  • RBI's SMA-0, SMA-1 and SMA-2 classes give a regulatory frame for the first 90 days of delinquency.
  • Roll rate by bucket is the core test of whether a collection strategy works.
  • AI-based risk scoring within each bucket stops lenders treating every 15-DPD account the same way.

What Is a Bucket-Wise Collection Strategy?

A bucket-wise collection strategy assigns collection effort by how far an account has fallen behind. It treats delinquency as a sequence of stages, each with its own cause, cost and best response.

A typical structure has six buckets: pre-due, 1 to 30 DPD, 31 to 60 DPD, 61 to 90 DPD, over 90 DPD and written-off. Many lenders then split each bucket by product, ticket size and risk score.

How Do DPD Buckets Map to RBI's SMA Classes?

DPD buckets map almost directly to RBI's SMA classes. According to RBI's NBFC Resolution of Stressed Assets Directions, 2025 (28 November 2025), SMA-0 covers accounts overdue up to 30 days, SMA-1 more than 30 and up to 60 days, and SMA-2 more than 60 and up to 90 days.

Under RBI's NBFC Income Recognition, Asset Classification and Provisioning (IRACP) Directions, 2025, an asset overdue for more than 90 days becomes a non-performing asset (NPA) for middle and upper layer NBFCs. Base layer NBFCs moved to the same 90-day norm from 31 March 2026.

Classification runs as part of the day-end process. So an account can move from SMA-2 to NPA overnight if the 91st day passes unpaid.

Why Do Uniform Collection Strategies Fail?

Uniform collection strategies fail for five reasons: they ignore why borrowers miss payments, waste expensive staff on self-curing accounts, reach high-risk accounts too late, use the wrong tone and never test what works. Each has a direct fix.

CauseEffectPrevention Check
Same calls for every overdue accountCost spent on accounts that would self-cureRisk score within each bucket
Field visits only after 60 DPDHigh-risk accounts roll before contactEarlier field triggers for high scores
One script for all stagesReminders sound like threats, or vice versaBucket-specific scripts
No tracking of roll ratesProblems seen only at NPAMonthly roll rate by bucket
Allocation by agent preferenceEasy accounts worked, hard ones ignoredSystem allocation with targets

Why Do Borrowers in Different Buckets Need Different Treatment?

Borrowers in different buckets need different treatment because the reasons for non-payment change with time. Early misses are often about timing, a failed debit or a forgotten date, while accounts past 60 DPD usually reflect real cash stress, disputes or intent.

That is why early buckets suit reminders and easy payment options. Later buckets need negotiation, restructuring checks or legal steps.

What Is the Right Bucket-Wise Collection Strategy Workflow?

The right bucket-wise collection strategy workflow has six steps: define buckets, score risk, assign channel and script, allocate staff, set escalation triggers and measure roll rates. Each step turns a static bucket into a managed queue.

  1. Define buckets and sub-segments by product and ticket size.
  2. Score risk within each bucket.
  3. Assign channel and script to each segment.
  4. Allocate staff and capacity.
  5. Set escalation triggers between buckets and channels.
  6. Measure roll rates and refine monthly.

Steps 1 and 2: How Should You Segment Within Each Bucket?

Segment within each bucket by risk, not just by DPD. A first-time defaulter with a bounced debit and a strong history is a different case from a habitual late payer at the same DPD.

Useful signals include payment history, bounce reason, contactability, ticket size and whether the loan is secured. Digiqt's collections prioritisation AI agent shows how accounts can be ranked by likely recovery.

Step 3: Which Channel and Script Suit Each Bucket?

Each bucket suits a different mix of channel, script and owner, moving from low-cost digital contact to personal and legal effort.

BucketPrimary ChannelScript FocusOwner
Pre-dueSMS, WhatsApp, emailReminder and payment linkAutomated
1–30 DPD (SMA-0)Digital plus tele-callingReason for miss, easy paymentEarly-stage tele team
31–60 DPD (SMA-1)Tele-calling, selective fieldCommitment and dateSenior tele team
61–90 DPD (SMA-2)Field visit, supervisor callResolution before NPAField team and supervisors
Over 90 DPDField, legal, settlementRestructuring, settlement or legalRecovery and legal teams

Payment friction is a common cause of early misses. Our guide to digital payment links for loan collections explains how to make paying easy at every bucket.

Step 4: How Should Staff Be Allocated Across Buckets?

Staff should be allocated by the volume and difficulty of each bucket, with senior people on the accounts where skill changes the outcome. Early buckets need many accounts per agent, while late buckets need fewer accounts and more time per case.

For field teams, plan routes around the highest-value SMA-2 accounts first. Our guide to the field collection app covers how field visits can be scheduled and recorded.

Step 5: When Should an Account Escalate?

An account should escalate when its risk changes, not only when it crosses a DPD boundary. Broken promises, unreachable numbers and a second bounce are all triggers to move a case to a senior agent or field visit early.

Unreachable borrowers need tracing before they need pressure. Our guide to skip tracing for loan collections explains lawful ways to re-establish contact.

All escalation must respect conduct rules. RBI's NBFC Responsible Business Conduct Directions, 2025 bar recovery calls before 8:00 a.m. and after 7:00 p.m., with separate rules for microfinance loans.

Step 6: How Do You Know a Bucket Strategy Is Working?

You know a bucket strategy is working when roll rates fall and cure rates rise bucket by bucket. Our guide to collection efficiency explains how to measure collections against what was due.

Test changes on a control group. Change one variable at a time, such as script or call timing, so you know what caused the improvement.

Do your 1–30 DPD accounts all get the same calls, whatever their risk?

Talk to Our AI Development Specialists

Visit Digiqt to see how risk scoring can shape the treatment for every DPD bucket.

How Much Does a Weak Bucket Strategy Cost a Lender?

A weak bucket strategy costs a lender higher roll rates, which push more outstanding into costlier buckets. In this illustrative scenario, 6,000 accounts enter the 1–30 DPD bucket each month, with an average outstanding of ₹80,000.

With a uniform approach, 25% roll into 31–60 DPD, which is 1,500 accounts and ₹12 crore of outstanding. With bucket-wise treatment and risk scoring, the roll rate falls to 18%, or 1,080 accounts and ₹8.64 crore.

Assume each account in the 31–60 bucket costs ₹600 a month to work through calls and visits.

MeasureUniform StrategyBucket-Wise Strategy
Accounts entering 1–30 DPD6,0006,000
Roll rate to 31–60 DPD25%18%
Accounts rolling1,5001,080
Outstanding rolling₹12 crore₹8.64 crore
Monthly cost to work rolled accounts₹9 lakh₹6.48 lakh

The lesson for leadership: every account stopped in an early bucket avoids higher collection cost later and reduces the pool that can reach NPA. Small roll-rate gains compound across every bucket.

Where Does Custom AI Agent Development Fit in Bucket-Wise Collection Strategy?

Custom AI agent development fits where lenders need risk scores, channel choices and escalation rules tuned to their own portfolio. Digiqt's custom AI agent development services include AI and data strategy, custom machine learning (ML) modelling, business process automation, explainable AI and analytics and reporting.

For bucket-wise collections, a custom AI agent can be built to:

  • Score each account within its bucket using repayment, bounce and contact data.
  • Recommend channel, timing and script for each segment.
  • Flag accounts likely to roll before they cross a bucket boundary.
  • Explain each score, so collections managers and auditors can see why an account was treated a certain way.

Digiqt's process runs from a Discovery and Business Understanding Sprint through feasibility, model selection and validation to deployment and continuous improvement. The early-stage collections AI agent and early delinquency warning AI agent show the kind of use cases involved.

Which Bucket Strategy Fits Your Portfolio?

The right bucket strategy depends on product, ticket size and borrower profile. Secured and small-ticket books need very different escalation paths.

PortfolioRecommended Approach
Small-ticket digital personal loansDigital-first through 30 DPD, tele to 90, settlement or online dispute resolution after
Vehicle loansEarlier field visits for high-risk accounts, repossession per policy
Microfinance loansCentre-based collection, microfinance-specific contact rules
MSME loansRelationship manager contact early, legal track after NPA
Home loansTele and branch contact, restructuring checks before legal action

Late-bucket accounts often need a negotiated exit. Our guide to one-time settlement for loans covers RBI's rules on compromise settlements.

Which Metrics Show a Bucket-Wise Collection Strategy Is Working?

Five metrics show whether a bucket-wise collection strategy is working: roll rate, cure rate, resolution rate, cost to collect and promise-kept rate. Track each one by bucket and segment.

MetricFormulaWhat a Weak Number Signals
Roll rateAccounts moving to the next bucket ÷ accounts in the bucketTreatment not stopping deterioration
Cure rateAccounts returning to current ÷ accounts in the bucketWeak early contact or payment friction
Resolution rateAccounts cured or stabilised ÷ accounts allocatedAllocation or script problems
Cost to collectCollection cost ÷ amount collected, by bucketExpensive channels on easy accounts
Promise-kept ratePromises to pay honoured ÷ promises madeWeak commitment scripts

Which Mistakes Weaken Bucket-Wise Collection Strategies?

Four mistakes weaken bucket-wise collection strategies, and each is a design choice.

  • Using DPD as the only segmentation. Risk differs widely inside each bucket.
  • Escalating only at bucket boundaries. Broken promises should trigger action immediately.
  • Measuring collections in total. Without roll rates by bucket, you cannot see where accounts slip.
  • Ignoring conduct rules in late buckets. Pressure rises with DPD, and so does complaint and regulatory risk.

What Is Changing Next?

RBI's NBFC Responsible Business Conduct Third Amendment Directions, 2026 (6 August 2026) take effect on 1 January 2027. They limit contact to between 8:00 a.m. and 7:00 p.m., require call recordings to be kept for six months and require recovery agents to hold an Indian Institute of Banking and Finance (IIBF) certificate.

Want a collection strategy that changes treatment as risk changes, not just as DPD grows?

Talk to Our AI Development Specialists

Visit Digiqt to see how custom AI agents can score, route and escalate collection accounts.

Sources

Read our latest blogs and research

Featured Resources

Technology

Collection Efficiency for Lenders: Formulas and Levers

Collection efficiency explained for lenders: formulas with worked arithmetic, bucket metrics, why public benchmarks mislead, and the levers that improve it.

Read more
Technology

Digital Payment Links for Loan Collections: One-Click EMIs

Digital payment links for loan collections: how to get overdue EMIs paid in one click, stay within RBI's direct-repayment rule and reconcile every payment fast.

Read more
Technology

Skip Tracing for Loan Collections: Reach Lost Borrowers

Skip tracing for loan collections: why borrowers go unreachable, which data you can lawfully use under DPDP and RBI rules, and how to find them faster.

Read more

About Us

We are a technology services company focused on enabling businesses to scale through AI-driven transformation. At the intersection of innovation, automation, and design, we help our clients rethink how technology can create real business value.

From AI-powered product development to intelligent automation and custom GenAI solutions, we bring deep technical expertise and a problem-solving mindset to every project. Whether you're a startup or an enterprise, we act as your technology partner, building scalable, future-ready solutions tailored to your industry.

Driven by curiosity and built on trust, we believe in turning complexity into clarity and ideas into impact.

Our key clients

Companies we are associated with

Life99
Edelweiss
Aura
Kotak Securities
Coverfox
Phyllo
Quantify Capital
ArtistOnGo
Unimon Energy

Our Offices

Ahmedabad

B-714, K P Epitome, near Dav International School, Makarba, Ahmedabad, Gujarat 380051

+91 99747 29554

Mumbai

C-20, G Block, WeWork, Enam Sambhav, Bandra-Kurla Complex, Mumbai, Maharashtra 400051

+91 99747 29554

Stockholm

Bäverbäcksgränd 10 12462 Bandhagen, Stockholm, Sweden.

+46 72789 9039

Malaysia

Level 23-1, Premier Suite One Mont Kiara, No 1, Jalan Kiara, Mont Kiara, 50480 Kuala Lumpur

Lewes

16192 Coastal Highway, Lewes, Delaware 19958, USA

software developers ahmedabad
ISO 9001:2015 Certified

Call us

Career: +91 90165 81674

Sales: +91 99747 29554

Email us

Career: hr@digiqt.com

Sales: hitul@digiqt.com

© Digiqt 2026, All Rights Reserved