Skip Tracing for Loan Collections: Reach Lost Borrowers
Last reviewed: November 2026
Skip tracing for loan collections is the process of locating and re-establishing contact with a borrower who can no longer be reached on the phone numbers and addresses the lender holds. Done well, it uses the lender's own records, lawfully obtained data and field verification to reach the right person without breaching privacy or conduct rules.
This guide is for heads of collections and CROs at NBFCs, banks and fintech lenders. It covers why borrowers go unreachable, which data you can lawfully use, and how to build a skip-tracing process that is fast and defensible.
Key Takeaways
- Under the DPDP Act, 2023, personal data can be processed only for a lawful purpose, with consent or for certain legitimate uses, and consent is limited to data needed for that purpose.
- RBI treats calling before 8:00 a.m. or after 7:00 p.m. for overdue loans, and intruding on the privacy of family, referees and friends, as harassment.
- RBI's digital lending rules bar lending apps from accessing contact lists and call logs.
- Most skip cases are ordinary contact changes, so your own data and a field visit solve more cases than external lists.
Why Is Skip Tracing a Leadership Problem?
Skip tracing is a leadership problem because an unreachable borrower cannot be helped, restructured or collected from, so the account keeps rolling into higher buckets. At the same time, aggressive tracing is where many lenders cross conduct and privacy lines.
RBI's NBFC Responsible Business Conduct Directions, 2025 prohibit intimidation and harassment in recovery. That includes acts that intrude on the privacy of a debtor's family members, referees and friends, threatening or anonymous calls, and false or misleading representations.
The Digital Personal Data Protection Act, 2023 adds a second test. Every piece of personal data used to trace a borrower needs a lawful basis, and penalties for some breaches reach ₹250 crore.
Why Do Borrowers Become Unreachable?
Borrowers usually become unreachable for six reasons: a changed number, a new job, a move, wrong data at onboarding, deliberate avoidance or death. Each needs a different response.
| Reason | What Happened | Where It Shows |
|---|---|---|
| Changed number | SIM replaced or number recycled | Number reaches a stranger or is switched off |
| New job | Employer changed, salary account moved | Mandate bounces, salary credits stop |
| Migration | Moved city for work | Field visit finds a locked or let-out house |
| Wrong onboarding data | Typo or agent-entered number | Unreachable from the first EMI |
| Deliberate avoidance | Borrower ignores known channels | Number active, calls never answered |
| Death or serious illness | Borrower cannot respond | Family reports at field visit |
Unreachable from the first EMI is a sourcing problem, not a collections problem. Feed it back to the onboarding team and the sourcing channel.
How Do Lenders Run Compliant Skip Tracing?
Lenders run compliant skip tracing in five layers, starting with their own data and moving outward only when needed. Each layer has a clear lawful basis.
- Data hygiene at onboarding: verify contact data and capture alternates with notice.
- Internal data first: search every system you already hold.
- Approved external data: use only sources your privacy lead has cleared.
- Field verification: visit the last known address and workplace.
- Escalation: move to legal notice when tracing is exhausted.
Layer 1: Can You Prevent Skips at Onboarding?
Verify the mobile number with a one-time password and capture an alternate number and email with a clear notice of purpose. Under the DPDP Act, consent must be specific and limited to data necessary for the stated purpose.
Ask borrowers to update contact details during servicing, for example when they log in or call. Small prompts prevent many skips later.
Layer 2: What Does Your Own Data Already Show?
Search every system you hold: application data, servicing records, call notes, payment channels and earlier loans. A borrower who paid once through a new UPI handle may have left a new contact trail in your own records.
This is also the most defensible layer. The data was collected by you, for this loan relationship.
Layer 3: Which External Sources Are Safe?
Use external data only where your privacy lead has confirmed a lawful basis and your agreements permit it. Never buy scraped contact lists or use phone contact harvesting.
RBI's digital lending rules require lending apps to desist from accessing contact lists, call logs and files, as explained in this guide to RBI digital lending guidelines compliance. Our DPDP Act compliance guide for lenders covers consent and purpose limits in more depth.
Layer 4: How Should Field Teams Trace Borrowers?
Send field teams to the last known address and workplace, with every visit logged and geo-tagged. A field collection app records visits, outcomes and new contact details as they are found.
Agents must identify themselves and must not mislead neighbours or colleagues about why they are asking. Disclosing the debt to third parties risks breaching RBI's privacy protections.
Layer 5: When Do You Escalate?
When tracing is exhausted, send legal notices to the address on record, following your board-approved recovery policy. Record every attempt, because the trail is your evidence of fair conduct.
Not every untraced account justifies the same legal spend. Digiqt's recovery rate prediction AI agent shows how expected recovery can be estimated account by account, so legal effort goes where it is most likely to pay back.
When you do reach a borrower, have an offer ready. Digiqt's payment plan optimisation AI agent shows how affordable repayment plans can be generated and ranked by likelihood of completion.
How many of your unreachable accounts have only been called, never traced?
Talk to Our Lending Technology Specialists
Visit Digiqt to see how custom AI agents can search your own data first and send field teams only where they are needed.
How Does Skip Tracing for Loan Collections Work on a Real Portfolio?
On a real portfolio, skip tracing for loan collections works best as a sequence, with each layer handling what the last one could not. In this illustrative scenario, an NBFC has 18,000 personal loan accounts at 31 to 90 days past due (DPD), and 2,700 of them, or 15%, cannot be reached.
The skip-tracing team works the 2,700 accounts through the layers.
| Layer | Accounts Reached | Running Total |
|---|---|---|
| Internal data: alternate numbers, emails, new payment trails | 1,100 | 1,100 |
| Field visit to last known address or workplace | 600 | 1,700 |
| Still unreachable, moved to legal notice | 1,000 | Not reached |
The lender re-establishes contact with 1,700 of 2,700 borrowers, about 63%. At an average overdue of ₹42,000, that puts ₹7.14 crore of overdues back into active conversation.
No external list was bought, and no relative was told about the debt. Every reach came from the lender's own data or a logged field visit.
The lesson for leadership: most skips were findable with data the lender already had. The bottleneck was searching it, not buying more.
Which Skip-Tracing Approach Fits Your Lending Model?
The right skip-tracing approach depends on your ticket size, borrower geography and field coverage. Small digital loans need automated internal search, while secured loans justify deeper field tracing.
| Situation | Recommended Approach |
|---|---|
| Small-ticket digital loans | Automated internal-data search, digital nudges to verified channels |
| Salaried personal loans | Workplace verification, updated contact capture at servicing |
| Secured loans with collateral | Field tracing at address and asset location |
| Migrant or rural borrowers | Local field partners, village-level address verification |
| Agency-outsourced collections | Approved data sources only, audited call and visit logs |
Can You Contact References and Relatives?
Treat references and relatives as a last resort, and never as pressure. RBI prohibits intruding on the privacy of a debtor's family members, referees and friends, and their numbers are their own personal data under the DPDP Act.
If your policy allows any contact, limit it to asking the person to pass on a message. Never disclose the loan or the amount.
Where Does Custom AI Agent Development Fit?
Skip tracing depends on your systems, data sources and policies, so it rarely fits a generic tool. Digiqt's custom AI agent development services include AI and data strategy, proof-of-concept and MVP builds, custom machine learning models, business process automation and explainable AI.
The development process runs from a discovery sprint and problem statement analysis through feasibility, model selection and dataset assessment, validation, and deployment with continuous improvement. For a skip-tracing agent, that means mapping every internal data source, encoding data-use rules and logging each search.
Prioritisation matters too. Digiqt's collections prioritisation AI agent shows how accounts can be ranked by recovery likelihood, with compliance limits such as contact frequency and consent checked before outreach.
For accounts that are reachable but slipping, Digiqt's early stage collections AI agent shows how the best contact channel and timing can be chosen for each borrower.
Which Metrics Show Skip Tracing Is Working?
Five metrics show whether skip tracing is working: skip rate, trace success rate, time to trace, right-party contact after trace and complaints from third parties. Track them by product, channel and agency.
| Metric | Formula | What a Weak Number Signals |
|---|---|---|
| Skip rate | Unreachable accounts ÷ delinquent accounts | Weak contact data at onboarding |
| Trace success rate | Accounts re-contacted ÷ skip accounts worked | Internal data is not being searched |
| Time to trace | Days from skip flag to re-contact | Accounts rolling to higher buckets meanwhile |
| Right-party contact after trace | Contacts with the actual borrower ÷ traced accounts | Wrong people being reached |
| Third-party complaints | Complaints from non-borrowers ÷ skip accounts worked | Privacy or conduct breaches |
Feed these into your collection efficiency reporting. Recovered contact is only valuable if it turns into payment.
Which Mistakes Make Skip Tracing Risky?
Four mistakes make skip tracing risky, and each can cost more than the overdue amount.
- Buying contact lists without checking the source. Data with no lawful basis creates DPDP exposure.
- Calling relatives as pressure. RBI explicitly protects the privacy of family, referees and friends.
- Agencies working without logs. Unlogged calls and visits cannot be defended in a complaint.
- Skipping internal search. Most new contact details are already in your own systems.
What Is Changing Next?
The DPDP Rules, 2025 were notified on 14 November 2025 with an 18-month phased rollout, so most obligations apply from May 2027. Consent notices, purpose records and erasure rules will then apply to every data source used in collections.
Collections supervision is also tightening around conduct. Lenders that can show a clean, logged and lawful trail for every trace will be better placed for both complaints and inspections.
Want a skip-tracing process that finds more borrowers and stands up to a DPDP audit?
Talk to Our Lending Technology Specialists
Visit Digiqt to see how our custom AI agent development team builds collections agents around your data and policies.
