Not every borrower responds to the same agent approach. A high-value, first-time defaulter with a temporary cash flow problem responds to an empathetic, senior agent who can discuss restructuring options — and may become a long-term customer if the interaction is handled well. A repeat defaulter who has shown persistent non-cooperation needs a different profile of agent: experienced with negotiation, familiar with legal escalation, and not easily deflected. The Field Collections Coordinator AI matches agent to borrower on four dimensions — not based on geography alone, but on the profile combination most likely to produce the outcome the situation requires.
Not every borrower responds to the same agent approach. A high-value, first-time defaulter with a temporary cash flow problem responds to an empathetic, senior agent who can discuss restructuring options — and may become a long-term customer if the interaction is handled well. A repeat defaulter who has shown persistent non-cooperation needs a different profile of agent: experienced with negotiation, familiar with legal escalation, and not easily deflected. The Field Collections Coordinator AI matches agent to borrower on four dimensions — not based on geography alone, but on the profile combination most likely to produce the outcome the situation requires.
Why arbitrary agent assignment costs collection outcomes
In most field collections teams, agent assignment is based on geographic territory — agent A covers Whitefield, agent B covers Koramangala, and the borrower's address determines which agent visits. This is administratively simple and operationally logical, but it ignores the probability of success. A borrower who is a first-time defaulter, who speaks only Kannada, who runs a small textile business, and who has a 90-day DPD on a ₹24 lakh loan, will have a materially different outcome with an agent who has textile sector knowledge and Kannada fluency than with an agent who has been handling unsecured personal loan defaulters and speaks primarily Hindi. The geographic assignment produces a visit. The profile match produces a collection.
The Field Collections Coordinator AI considers four matching dimensions: borrower profile (DPD bucket, loan type, ticket size, prior contact history, stated reason for default), required outcome (payment collection, PTP commitment, restructuring discussion, legal notice delivery), agent capability (language, sector knowledge, experience with this DPD bucket, recent success rate with similar profiles), and escalation requirements (whether the account's contact history requires a two-agent team or a specific seniority level for authority to discuss settlement terms).
"An agent who is expert at negotiating with repeat personal loan defaulters is not the right agent for a first-time business loan defaulter having a cash flow problem. Profile matching produces outcomes that geographic matching does not."
The agent-borrower matching matrix: four high-value accounts
Account: Rajan Textiles Pvt Ltd — LA-2024-4821 · DPD: 91 · ₹28.4L MSME
Agent match: 91/100 · R. Subramaniam assigned
Borrower profile
SectorTextile manufacturing · Tirupur supply chain
DPD bucket91 days — first NPA threshold
Prior default historyNone — first default event
Stated reasonDelayed receivables from major buyer
Language preferenceKannada, Tamil (some English)
Required outcomeRestructuring discussion + PTP
Agent R. Subramaniam — why matched
Sector knowledgeTextile / garment — 4 years
LanguagesKannada, Tamil, Hindi
DPD 91–120 success rate68% PTP rate (last 90 days)
First-default empathy scoreHigh — rated by supervisor
Restructuring authorityApproved to discuss up to 3-month restructuring
Geographic availabilityRajajinagar territory · 8km from account
Rajan Textiles is a first-time defaulter in a sector-specific cash flow squeeze — this is an account to recover with empathy, not pressure. Subramaniam's textile sector knowledge allows a peer conversation about the business situation. His restructuring authority means he can offer a genuine path forward in the first visit, not a "I'll pass your case upward" response that produces no outcome.
Account: Suresh Reddy — LA-2023-7712 · DPD: 284 · ₹8.2L Personal Loan
Agent match: 78/100 · P. Anjaneyulu assigned · 2-agent team
Borrower profile
DPD bucket284 days — Doubtful D1
Prior default history2 prior personal loans — 1 settled, 1 written off
Contact historyVerbal aggression at last 2 visits · Refused to meet
Last PTP3 PTPs — all broken
Language preferenceTelugu, some Kannada
Required outcomeSettlement discussion or legal notice delivery
Agent P. Anjaneyulu — why matched
DPD 200+ experience7 years · highest team specialist
LanguagesTelugu, Kannada, Hindi
Legal notice authorityCertified to serve Section 13(2) notices
Volatile borrower scoreExperienced (14 volatile accounts, no incidents)
2-agent teamPaired with K. Manjunath (proximity available)
Settlement authorityAuthorised up to 20% principal reduction
Suresh Reddy's profile — multiple prior defaults, broken PTPs, verbal aggression — requires an experienced agent with legal authority and a paired partner for safety. Anjaneyulu's 7-year DPD 200+ experience means he will not be deflected by aggressive behaviour, and his settlement authority gives him genuine leverage to produce an outcome rather than another broken PTP.
Account: Dr Priya Venkatesh — LA-2024-9012 · DPD: 45 · ₹65L Home Loan
Agent match: 88/100 · S. Nandakumar assigned · Senior
Borrower profile
DPD bucket45 days — early bucket, high value
Loan typeHome loan · ₹65L · doctor, private practice
Prior contactPhone — co-operative, cited EMI oversight
Risk of escalationLow — genuine administrative miss
LanguageEnglish, Tamil
Required outcomePayment recovery · Preserve relationship
Agent S. Nandakumar — why matched
Professional borrower experienceDoctors, CAs, corporate salaried — specialist
LanguagesEnglish, Tamil, Kannada
Early-bucket success rate82% same-visit payment (last 90 days)
Relationship preservation ratingHigh — borrower satisfaction scores
Ticket size experienceHandles ₹50L+ accounts regularly
SenioritySenior agent — appropriate for high-value borrower
A ₹65L home loan borrower at 45 DPD for an administrative oversight is an account to recover and retain — a heavy-handed approach will damage the relationship permanently. Nandakumar's 82% same-visit payment rate on early-bucket accounts and his professional borrower experience make him the right agent to produce payment and preserve the customer.
4Matching dimensions — borrower profile, required outcome, agent capability, escalation requirement
91Rajan Textiles match score — textile sector knowledge, Kannada fluency, restructuring authority, first-default empathy
2-agentSuresh Reddy assigned a paired team — 3 prior broken PTPs and documented verbal aggression require it
82%Nandakumar's same-visit payment rate for early-bucket professional borrowers — the match produces the outcome
The right agent is not the nearest agent — it is the agent most likely to produce the right outcome
Geographic assignment is an efficiency decision. Profile matching is an outcomes decision. Both matter — but an agent who is nearby but wrong for the borrower profile produces a wasted visit as surely as an agent who is too far away. The Field Collections Coordinator AI optimises for both: the agent assigned to each account is the one whose capability profile best matches what the account requires — and that agent's route is then optimised to minimise travel. The assignment produces the outcome. The route makes the assignment possible at scale.
