Loan restructuring in India is historically slow — a borrower in financial hardship must call the lender, get through to the right department, explain their situation repeatedly to multiple agents, and wait for a credit committee decision while DPD accrues and CIBIL deteriorates. Kallix compresses this process by automating the intake and option-presentation stages, so the human credit officer only needs to make and document the final decision.

The AI's role in the restructuring workflow:
1. **Financial data collection**: structured interview to capture current income, employment status, other loan EMIs (FOIR calculation), and reason for hardship — feeds the lender's restructuring eligibility engine.
2. **Restructuring option presentation**: the AI reads the lender's pre-approved restructuring matrix (configured by product, DPD, outstanding, and hardship type) and presents 2–3 applicable options clearly, with monthly EMI amounts and total cost differences stated.
3. **KFS dispatch**: immediately on expression of interest in a specific option, the KFS (Key Facts Statement — interest rate, tenure, revised EMI, total cost of credit, processing fee) is dispatched via WhatsApp/email within 60 seconds per RBI DLG 2022.
4. **Cooling-off acknowledgement**: the AI confirms the borrower has 3 days to review and withdraw from the restructuring offer without penalty.
5. **Documentation trigger**: when the borrower confirms acceptance after the cooling-off window, the AI triggers the loan modification agreement generation and routes to the credit officer for execution.
6. **NACH amendment**: new EMI amount and due date flagged to the mandate management team for NACH re-registration.

Kallix customers in retail and MSME lending report that AI-assisted restructuring intake reduces credit officer time per case by 65–75% — officers review pre-packaged cases rather than conducting full intake interviews.

The restructuring option that works best for a borrower depends on the nature and duration of their hardship — a 2-month salary delay needs a different solution than a business closure. Kallix's restructuring presentation logic is driven by the borrower's hardship classification, DPD stage, and product type.

**Type 1 — EMI Deferral (1–3 months)**:
Best for: short-term cash flow disruption — salary delay, medical expense, seasonal business dip. Mechanism: 1–3 EMIs skipped; accrued interest and principal added to the remaining EMIs or the tenor extended by the deferred months. CIBIL impact: none if executed before 30 DPD; the account is treated as 'rescheduled' not 'restructured' within the lender's standard product terms in many cases. Cost to borrower: modest — primarily the additional interest on deferred amount.

**Type 2 — Tenure Extension**:
Best for: permanent income reduction — job downgrade, reduced business income. Mechanism: original EMI reduced by extending the loan tenure. Example: Rs 22,000/month EMI on a 5-year personal loan becomes Rs 15,500/month on a 7-year extension. CIBIL impact: reported as 'restructured' if the extension is beyond original terms. RBI requires the extended account to be monitored for at least 12 months before re-upgrading to standard.

**Type 3 — Step-Down EMI**:
Best for: contractual income disruption — notice period, business seasonality, interim project. Mechanism: EMI reduced to 50–70% of original for 3–6 months, then steps back to original or slightly above. Total loan cost increases due to additional interest accrual during the step-down phase.

**Type 4 — Interest Moratorium**:
Best for: acute temporary hardship — medical, natural disaster, crop failure. Mechanism: interest charges are paused; principal repayments may continue. Less common in retail lending; more frequent in agricultural and MSME lending. RBI's COVID-19 moratorium (March–August 2020) was the largest deployment of this type in India.

**Type 5 — OTS (One-Time Settlement)**:
Best for: accounts where full repayment is not viable — 60+ DPD, NPA-approaching. Mechanism: lender accepts discounted settlement. CIBIL impact: 'Settled' status for 7 years — detailed in a separate question.

India's restructuring regulatory landscape has evolved significantly since the pre-IBC era. Kallix's compliance configuration is built around the current operative frameworks.

**RBI Prudential Framework for Resolution of Stressed Assets (June 2019)**: The primary framework for large-exposure restructuring. For retail and MSME lending below Rs 100 crore, the requirements are less prescriptive — lenders have policy discretion on restructuring terms but must avoid classification benefits if the restructuring is driven by financial difficulty rather than contractual variation.

**Classification implications**:
- A loan restructured before NPA classification (before 90 DPD) with a viable resolution plan can remain a Standard Asset, subject to additional provisioning (typically 5% additional provision on the restructured standard asset).
- A loan restructured after NPA classification is 'upgraded' only after 12 months of satisfactory performance post-restructuring — lenders may not upgrade immediately on restructuring.
- For NBFCs, the NBFC Master Directions 2022 govern restructuring classification — thresholds and upgrade criteria differ slightly from scheduled commercial banks.

**RBI Digital Lending Guidelines 2022 compliance for restructuring**:
- KFS delivery within 60 seconds
- 3-day cooling-off period for new restructured loan terms
- No automatic enhancement of outstanding credit
- LSP (Kallix) must be disclosed at point of restructuring discussion
- Grievance officer pathway provided at point of offer

Kallix's compliance guardrails:
1. KFS dispatch is non-negotiable — the AI cannot proceed past option presentation without confirming KFS was sent.
2. Cooling-off period is tracked in the LMS — acceptance before 3 days triggers an alert.
3. Credit enhancement prohibition: the AI will not present any option that increases the borrower's principal outstanding beyond what was already owed (capitalised interest is permitted; new disbursement is not).
4. All restructuring interactions are recorded and tagged as 'restructuring call' in the compliance dashboard.

The quality of a restructuring decision depends on the quality of the financial data gathered. A generic 'I can't pay' results in a generic offer; a structured income-expense-obligation profile enables the lender to calculate a revised FOIR and offer an EMI the borrower can actually sustain — which is the only restructuring that works.

Kallix's financial intake structure:

**Retail borrower intake** (personal loan, home loan, vehicle loan, credit card):
1. Monthly take-home income: 'What is your current monthly take-home salary or income? Include all sources — salary, rental income, business income.'
2. Employment status: still employed (same employer/changed employer), self-employed, unemployed, business paused.
3. Household expenses: 'Approximately how much do you spend monthly on household necessities — food, utilities, school fees, rent?' (Rough estimate — this calibrates the available surplus.)
4. Other EMIs: 'Apart from this loan, what are your total monthly loan repayments across all lenders?' FOIR is calculated on these inputs.
5. Reason and duration: 'What has caused the payment difficulty, and how long do you expect this situation to last?' This determines deferral vs tenure extension vs OTS pathway.

**MSME borrower intake** (business loan, MSME term loan, CC/OD account):
- Above fields plus: monthly business turnover (approximate), last GST filing quarter's output tax, number of employees, nature of cash flow disruption (buyer payment delay, inventory lock-up, demand slowdown).
- Account Aggregator consent: for MSME borrowers, Kallix integrates with the AA framework (GSTIN consent + bank statement pull) to auto-populate financial data rather than relying solely on verbal declaration — reduces manipulation and speeds credit officer review.

**Data output to LMS**: Within 15 minutes of call end, the financial intake is structured into the lender's restructuring eligibility template. The credit officer receives: borrower hardship profile, calculated current FOIR, suggested restructuring path (from the matrix), and the AI-captured income/expense data — all pre-loaded before the officer's follow-up call.

CIBIL and other bureau reporting for restructured accounts is governed by RBI Master Direction on Credit Information Companies and the specific reporting instructions that accompany RBI's restructuring circulars. The AI explains these nuances clearly — without exaggerating risk or minimising it.

**Scenario 1 — Pre-30 DPD administrative deferral** (e.g., lender extends due date by 30 days within product terms): If the lender treats this as a contractual variation rather than a stressed restructuring, bureau reporting may show no change. This depends entirely on the lender's reporting policy — Kallix does not guarantee bureau outcome; it advises the borrower to confirm with the lender's credit reporting team.

**Scenario 2 — Formal restructuring (tenure extension, step-down EMI)**: Reported to bureaus with an account status flag — 'Restructured' or 'Rescheduled' depending on the severity and the lender's reporting classification. This flag is visible to future lenders during credit assessment. Practical impact: most lenders do not automatically reject a borrower with a 'Restructured' flag (unlike 'NPA' or 'Settled'), but they may apply additional scrutiny or charge a risk premium of 50–100 bps.

**Scenario 3 — OTS**: Reported as 'Settled' — the most negative classification short of a written-off NPA. Visible for 7 years. Future credit is significantly impaired.

**What the AI communicates**: 'Restructuring your loan will be recorded on your credit report. This is less severe than a default or settlement, but future lenders can see it. If you are planning to apply for a home loan or large credit facility in the next 2–3 years, discuss with our credit team whether there is a path to resolve without a restructuring flag.'

For borrowers who are in early DPD (31–45 DPD) and have the capacity to fully repay with a short extension, the AI presents the option to bring the account fully current — which avoids the restructuring flag entirely — before presenting formal restructuring alternatives.

The KFS requirement is one of the most operationally demanding provisions of the RBI DLG 2022 — 60 seconds is a very tight window that manual processes routinely miss. Kallix automates this entirely.

KFS content for a restructured loan (per RBI DLG 2022 mandatory 8-point disclosure):
1. **Annual Percentage Rate (APR)**: the full effective cost including processing fees, insurance, and any other charges — not just the nominal interest rate.
2. **Revised EMI amount**
3. **Revised number of EMIs (tenure)**
4. **Total amount payable over the revised loan life**
5. **Cooling-off period**: 3 days from KFS receipt to cancel without penalty
6. **Grievance officer details**: name, designation, phone, email
7. **LSP details**: Kallix's name and role as the technology service provider
8. **Recovery agent contact**: if collections are being conducted by a separate agent

KFS dispatch mechanism: Kallix generates the KFS document dynamically from the LMS — it pulls the current outstanding, the restructuring terms being offered, and the lender's fee schedule to compute the revised APR in real time. The document is formatted as a PDF and dispatched to:
- Registered WhatsApp number (highest read rate, typically >90% within 5 minutes)
- Registered email (required by RBI even if WhatsApp is primary)

If dispatch fails (WhatsApp message undelivered, email bounces), the AI notes the failure, triggers a retry after 2 minutes, and flags the account for manual KFS delivery if both retries fail. The restructuring offer is paused until KFS delivery is confirmed — the AI will not proceed to acceptance capture without a confirmed KFS delivery timestamp in the LMS.

KFS receipt confirmation: the borrower is asked to confirm receipt ('Have you received the document on your WhatsApp?'). Verbal confirmation is recorded. For high-value restructuring (above Rs 10 lakh outstanding), a digital acknowledgement (WhatsApp reply button or email reply) is required before acceptance is processed.

The cooling-off provision is designed to prevent rushed decisions on restructured loan terms — which are complex financial products that affect the borrower's obligations for years. The AI operationalises it as a process checkpoint, not just a disclosure.

Cooling-off workflow:
1. **At option presentation**: 'You have 3 full days from today to review this offer. If you change your mind for any reason, you can cancel and we'll revert to the current loan terms with no penalty.'
2. **KFS delivery timestamp**: recorded in LMS as the cooling-off clock start.
3. **Day 1 check**: no acceptance processed for 24 hours regardless of borrower urgency.
4. **Day 2 reminder**: WhatsApp message sent to borrower: 'This is a reminder that your loan modification offer expires on [date]. You have until then to cancel without charge. If you'd like to accept or have questions, call us at [number].'
5. **Day 3 — acceptance window opens**: the AI's acceptance processing is enabled. If the borrower calls before Day 3 wanting to accept immediately, the AI explains: 'Your acceptance can be processed from [date] — this is to ensure you have had time to review the terms fully. I'll schedule a callback on [date] to confirm your decision.'
6. **Cancellation during cooling-off**: If the borrower cancels within 3 days, the LMS reverts the account to the original terms, no fees are charged, and a cancellation confirmation is sent. The DPD continues accruing from the original overdue date — the cooling-off pause does not freeze the DPD clock.

RBI does not mandate a minimum cooling-off period beyond 3 days, but Kallix recommends lenders offer 5 days for accounts above Rs 5 lakh outstanding, as this reduces cooling-off period cancellations (borrowers who cancel often do so because they felt rushed).

NACH mandate amendment is one of the most operationally complex steps in loan restructuring — it requires coordination between the LMS, the borrower, the borrower's bank, and NPCI's NACH registry. Delays here cause the first post-restructuring EMI to bounce, which re-triggers collection calls and creates borrower confusion. Kallix automates the entire mandate amendment flow.

NACH amendment workflow:
1. **Old mandate cancellation**: at loan modification execution, Kallix triggers the mandate cancellation instruction to the presenting bank via NPCI NACH portal. Cancellation takes 1–3 business days.
2. **New mandate registration link**: sent to borrower via WhatsApp within 30 minutes of loan modification execution. Two options: (a) eSign mandate — borrower completes digitally via Aadhaar OTP in 3–5 minutes; (b) physical NACH form — printed, signed, and submitted via bank branch (used for borrowers without Aadhaar-linked mobile).
3. **Registration timeline**: NPCI NACH registration takes 7–10 business days for physical mandate; 1–2 business days for eSign via NPCI's API-based registration.
4. **First EMI bridge**: The AI advises the borrower: 'Your first EMI under the new terms will be on [revised due date]. Since the mandate registration takes a few days, please also keep the option of a manual UPI payment ready for the first month, in case the mandate isn't yet active.' A UPI collect request is pre-prepared and sent 3 days before the first EMI date as backup.
5. **Mandate confirmation**: when NPCI confirms mandate registration, LMS updates with new UMRN (Unique Mandate Reference Number) and the old mandate UMRN is archived.
6. **Reminder for incomplete mandates**: if the borrower has not completed the mandate registration 5 days after the link was sent, the AI makes a follow-up call to assist with the eSign process or schedule a branch visit.

The moratorium interest communication failure during the RBI COVID-19 moratorium (March–August 2020) was a significant learning moment for the Indian banking sector — millions of borrowers took the moratorium believing it was interest-free, then discovered their outstanding had grown substantially. Kallix's restructuring scripts are designed to prevent this confusion.

Moratorium interest communication framework:

**What the AI states upfront**: 'During a 3-month moratorium, your EMI payments are paused. However, interest continues to accrue at your current rate of [X]% per month on the outstanding principal of Rs [Y]. Over 3 months, this comes to approximately Rs [Z]. This amount will be [added to your remaining EMIs / repaid as a lump sum in month 4 / spread across the tenure extension].'

**Three handling methods and their implications**:
1. **Capitalisation**: accumulated moratorium interest added to principal. All future EMIs are slightly higher. Most common method. Net effect: the loan costs more over its life, and if the tenure is not extended, EMIs increase.
2. **EMI enhancement**: current EMI continues, but the moratorium interest is spread across remaining months as a small EMI top-up. Less disruptive than lump sum.
3. **Lump sum at moratorium end**: borrower pays accumulated interest in full before regular EMIs resume. Least popular — borrowers in hardship typically cannot pay a lump sum at moratorium end.

**Comparison with no-moratorium option**: The AI always presents the total additional cost of the moratorium in rupees vs continuing to pay: 'If you can pay Rs 8,000 now (instead of your full EMI of Rs 18,500), we can keep the account current without a moratorium, and your total additional cost would be Rs 2,100 in late charges vs Rs 12,450 in moratorium interest. Which would you prefer?'

This comparison often leads borrowers in partial-hardship to choose a partial payment over a full moratorium — better for both the borrower and the lender.

The tendency in traditional collections is to lead with the lower EMI amount — 'your EMI drops to Rs 15,500' — without mentioning that the borrower pays Rs 1.18 lakh more over the loan life. This creates short-term agreement but long-term resentment and re-default when the borrower realises the true cost. Kallix's approach is full-cost transparency, which aligns with RBI DLG 2022's mandatory total cost of credit disclosure in the KFS.

Total cost calculation the AI presents for each option:

**Tenure extension example**:
- Current: Rs 22,000/month × 36 remaining months = Rs 7,92,000 total remaining payment
- Proposed: Rs 15,500/month × 60 months = Rs 9,30,000 total payment
- Additional cost: Rs 1,38,000
- AI states: 'The convenience of a lower EMI costs Rs 1,38,000 in additional interest over the extension period.'

**Step-down EMI example**:
- Step-down phase: Rs 11,000/month for 6 months = Rs 66,000
- Post step-up: Rs 23,200/month for 30 months = Rs 6,96,000
- vs current: Rs 22,000/month for 36 months = Rs 7,92,000
- Additional cost vs no restructuring: Rs 30,000 (the interest that accrues during the step-down phase)

**Why this transparency works commercially**: Borrowers who understand the full cost often elect a shorter restructuring window than they initially requested — 'If I could manage Rs 17,000 instead of Rs 15,500, would I save significantly?' The AI can instantly compute: 'Yes — Rs 17,000/month for 48 months reduces your additional interest to Rs 54,000 vs Rs 1,38,000 at Rs 15,500.' This collaborative optimisation increases borrower commitment to the agreed terms and reduces re-default risk.

Post-restructuring monitoring is a regulatory requirement — RBI's Prudential Framework requires that restructured accounts be monitored for at least 12 months before classification upgrade. Beyond compliance, monitoring is commercially essential: a restructured account that re-defaults quickly is worse than never having been restructured — the lender has given up collection leverage and still faces an NPA.

Kallix's 12-month monitoring programme:

**Month 1–3 (high-touch phase)**:
- Pre-debit reminder call 3 days before each EMI (same as standard reminder but flagged as 'restructured account — enhanced monitoring')
- EMI success or failure notification to the credit officer's dashboard within 2 hours of due date
- WhatsApp check-in at Month 2: 'How are you finding the revised EMI? Any upcoming changes to your income we should be aware of?'

**Month 4–12 (standard monitoring with quarterly wellness)**:
- Standard pre-debit reminders
- Quarterly financial wellness call: brief 3–4 minute check-in on employment status, income change, other obligations. The purpose is early warning — if the borrower's income has improved, flag for rate renegotiation; if deteriorated further, flag for pre-emptive re-restructuring before the next default.
- Post-restructuring DPD tracking: any missed EMI in the monitoring period is flagged as 'post-restructuring delinquency' in the LMS — a separate risk flag from standard DPD.

**Re-default protocol**:
- First missed post-restructuring EMI (1 DPD): immediate AI collection call + credit officer alert
- Second missed post-restructuring EMI (within 12 months): escalated directly to human collections head — standard AI soft collection protocol does not apply for repeat restructured defaults
- Third miss: account reviewed for NPA reclassification and legal referral

**Classification upgrade at 12 months**: if the account has paid all restructured EMIs on time for 12 consecutive months, the credit officer is prompted to initiate the standard asset upgrade in the LMS — removing the 'Restructured' classification and the additional provisioning requirement.

Re-default on a restructured loan is a more severe signal than a first-time default — it indicates the restructuring either did not adequately address the borrower's financial situation, or the borrower's circumstances have deteriorated further, or the borrower took the restructuring as a breathing room rather than a genuine resolution commitment.

Kallix's re-default protocol:

**First post-restructuring missed EMI**:
- The AI collection call is made within 2–4 hours, with a modified script that acknowledges the restructuring: 'I'm calling regarding your restructured loan account — your first EMI under the new terms on [date] has not been received. Has something changed since we set up the new arrangement?' The tone is empathetic but urgent — the credit officer is simultaneously notified.
- If the borrower gives a credible reason (NACH not yet active — common in month 1), the AI accepts this and schedules a manual payment reminder. If the reason is financial, the credit officer reviews within 24 hours.

**Second post-restructuring missed EMI within 12 months**:
- Bypasses AI — escalated directly to senior human collector.
- LMS flag: 'Restructured — re-default. Second restructuring review required.'
- Credit officer initiates a fresh financial assessment.

**Second restructuring**: Permissible under RBI Prudential Framework in exceptional cases (a fresh exogenous event — e.g., a second COVID wave, a natural disaster — distinct from the original hardship). However, a second restructuring triggers automatic NPA classification at the time of the second restructuring for bank accounts. For NBFCs, the treatment is similar but governed by NBFC Master Directions. This is a significant deterrent — the AI explains this clearly before a second restructuring is proposed.

**Write-off pathway**: An account that re-defaults and cannot sustain any restructured terms is reviewed for OTS or NPA classification and provisioning. AI supports by providing the human negotiator with full interaction history, both pre- and post-restructuring financial data, and PTP history.

Small-ticket account restructuring through the standard credit officer workflow is economically unviable — the cost of a full financial intake, credit committee review, and documentation for a Rs 1.5 lakh personal loan restructuring often exceeds the recovery value difference between restructuring and writing the account off. Green channel restructuring solves this by pre-approving a limited set of terms for eligible small accounts.

Green channel eligibility (lender-configured, typical parameters):
- Outstanding below Rs 2 lakh (adjustable — some lenders use Rs 5 lakh threshold for green channel)
- 31–60 DPD (not yet NPA-approaching)
- Clean payment history for first 6–12 months of loan life (first-time defaulter only)
- No prior restructuring on this account
- No fraud flag, no dispute flag, no distress flag

Green channel offer: a single pre-approved offer — not a menu of options. Example: '2-month deferral with interest capitalisation, no fee, account reverts to original terms after deferral.' The simplicity is deliberate — presenting too many options increases drop-off for low-ticket accounts.

End-to-end digital flow:
1. AI call: states the overdue, presents the green channel offer, confirms the borrower's interest.
2. KFS sent via WhatsApp (auto-generated from standard template — no dynamic calculation needed for pre-approved terms).
3. Cooling-off: 3-day window.
4. Acceptance: WhatsApp reply button or verbal confirmation on AI follow-up call.
5. NACH amendment: automated for deferral — no new mandate needed; the existing mandate is paused in the NPCI NACH system for 2 months and then reactivated.
6. Documentation: digital loan modification agreement sent via WhatsApp for eSign. No branch visit.

Green channel throughput: entire process from AI call to executed modification in 3–5 days. Standard channel: 12–18 days. Lenders processing 5,000+ small-ticket restructurings per month save an estimated Rs 85–1,200 per case in operational cost.

MSME restructuring has been given specific regulatory support by RBI beyond the standard stressed asset framework — recognising that small business cash flows are inherently more volatile than salaried income and that standard retail restructuring timelines are often too slow to prevent SME defaults.

RBI MSME restructuring permissions:
- **One-time MSME restructuring (RBI January 2019)**: Standard asset MSME loans up to Rs 25 crore outstanding could be restructured once without NPA classification, provided the borrower has a valid GST registration and the account was standard as on January 1, 2019. While this specific scheme has been superseded, it established the principle that MSME restructuring can occur without automatic classification downgrade — subsequent RBI circulars (COVID restructuring 2020, Resolution Framework 2021) extended this principle.
- **Credit Guarantee Scheme for Restructuring**: CGTMSE-backed loans have specific restructuring provisions — the lender must notify CGTMSE before restructuring a guaranteed loan, as restructuring affects the guarantee eligibility.

Kallix MSME restructuring intake:
- Account Aggregator consent for GSTIN + bank statement pull — reduces intake from 45–60 minutes to 8–12 minutes
- Business disruption classification: buyer concentration risk, input cost spike, demand slowdown, regulatory/license issue, working capital crunch
- Seasonality modelling: for businesses with seasonal cash flows (agriculture, retail, hospitality), the AI captures the expected recovery month and structures repayment to align with peak collection season
- Promoter guarantees: many MSME loans have promoter personal guarantees — if the promoter's personal financial position is sound even though the business is stressed, the AI routes to a collateral-secured repayment plan rather than pure restructuring

Post-restructuring MSME monitoring: quarterly GST filing confirmation + annual Account Aggregator pull to verify business is active and financial position is within restructuring assumptions.

Restructuring automation depends on more integration touchpoints than any other collections use case — the AI needs to read complex loan data, write back restructured terms, trigger documentation, and coordinate mandate changes across three systems simultaneously. Kallix's integration architecture for restructuring:

**LMS read operations**:
- Current outstanding (principal, accrued interest, pending charges)
- Original loan terms: disbursement date, original rate, original tenure, EMI amount
- Payment history: all EMIs paid, missed, and deferred in chronological order
- Existing restructuring history: prior deferrals, moratoriums, or modifications
- OTS eligibility flag and pre-approved OTS amount (if applicable)
- CGTMSE/guarantee flag for MSME accounts

**LMS write operations**:
- Restructuring intake data: borrower's financial profile, hardship classification, income/FOIR
- Restructuring option selected: type, new EMI, new tenure, effective date
- KFS dispatch timestamp and delivery confirmation
- Cooling-off start and end dates
- Acceptance timestamp and confirmation method (verbal/digital)

**Document management integration**:
- Loan Modification Agreement (LMA) auto-generated from LMS data + restructuring terms
- KFS document generated from LMS data + lender's fee schedule
- eSign via Leegality (API-based), DigiLocker (for Aadhaar eSign), or lender's own eSign platform
- Document delivery via WhatsApp (Kallix's BSP partner) + email + LMS document vault

**NACH management**:
- Old mandate cancellation instruction via NPCI NACH API or lender's mandate management system
- New mandate registration link generated and sent
- UMRN update in LMS on NPCI confirmation

**Integration deployment**: 4–6 weeks for LMS + document + NACH integration. eSign integration adds 1–2 weeks. Full UAT requires 200+ test cases covering all restructuring types, documentation flows, and mandate amendment scenarios.

The ROI case for AI-assisted restructuring operates across three dimensions: cost, speed, and outcome quality.

**Cost per case**:
- Manual: collector interview (30–45 min) + credit officer review (20–30 min) + documentation team (45–60 min) + NACH amendment coordination (30–45 min) = Rs 1,800–3,200 in fully-loaded human cost
- AI-assisted: 8–12 min AI intake call + 5–10 min credit officer review + automated documentation + automated NACH = Rs 350–600
- Cost saving per case: Rs 1,200–2,600

**Speed**:
- Manual: 12–18 days (intake queue 2–3 days + credit committee 3–5 days + documentation 3–5 days + NACH registration 7–10 days — with significant overlap in practice)
- AI-assisted: 3–5 days (AI intake same day + credit review 1 day + eSign same day + NACH registration 2–3 days)
- Speed matters because DPD accrues during the restructuring process: every day saved reduces CIBIL damage to the borrower and provisioning pressure on the lender

**Take-up rate** (percentage of eligible accounts that complete restructuring):
- Manual outreach: 42–55% (borrowers don't call in; branch visits are inconvenient; long queue times lead to abandonment)
- AI-assisted: 68–76% (AI calls the borrower proactively; digital process is completed in a single flow without branch visits; follow-up for cooling-off is automated)

**12-month portfolio quality**:
- 58–68% of AI-assisted restructured accounts remain Standard Assets at 12 months — compared to 44–52% for manually processed restructuring (Kallix production benchmark, retail lending cohort)
- Improvement driver: AI collects more complete financial data, leading to better-calibrated EMI offers that borrowers can actually sustain

Multi-product borrowers represent a disproportionate share of distressed retail loan portfolios — a borrower managing three EMIs simultaneously is more vulnerable to a single income disruption. Kallix's consolidated restructuring intake is designed for this profile.

Consolidated intake for multi-product borrowers:
1. The AI identifies all overdue or near-delinquent accounts at call start (LMS query by borrower PAN or mobile)
2. The call begins: 'I'm calling from [Lender Name] regarding your accounts. I see you have a personal loan account ending [XXXX] and a home loan account ending [YYYY]. I'd like to understand your situation and see how we can help with both. Is that okay?'
3. A single income/expense/FOIR intake is conducted — not a separate intake per product. The combined FOIR across all products is calculated once.
4. The AI presents restructuring options per product: 'For your personal loan, given your current income, a 12-month tenure extension reduces your EMI from Rs 18,500 to Rs 12,800. For your home loan, a 2-month deferral brings the account current without a restructuring flag. Would you like to explore both?'
5. Product-specific KFS documents are dispatched simultaneously — separate KFS per product as required by RBI DLG 2022, but delivered in a single WhatsApp message bundle.
6. Cooling-off periods run in parallel — both products have the same 3-day window from the same delivery timestamp.

Credit officer benefit: the combined restructuring package is presented as a single case in the LMS — one credit review covers both products, reducing approval cycle time by 40–50% vs two separate reviews.

Priority logic: if the borrower has capacity to fully resolve one product (e.g., home loan at 22 DPD can be brought current with a one-time payment) while restructuring the other (personal loan at 45 DPD), the AI presents this path first — resolving the easier account removes the bureau impact of the home loan and improves the borrower's credit position for the personal loan restructuring discussion.

Restructuring deployment is the most integration-heavy of all Kallix's collections modules — it touches LMS, document management, eSign, NACH mandate systems, and the Account Aggregator framework for MSME. The deployment phases:

**Phase 1 — Integration Setup (weeks 1–2)**:
- LMS API: read current loan data + write restructuring intake and new terms
- Restructuring matrix upload: lender's credit team provides eligible options by DPD, product, outstanding, and hardship type in a Kallix-provided template
- KFS template setup: lender's legal team reviews and approves KFS template for each product type
- eSign platform integration: API credentials for Leegality or DigiLocker
- NACH mandate management integration: old mandate cancellation + new registration flow
- Account Aggregator integration (for MSME): FIP consent flow setup with AA provider

**Phase 2 — Script & Compliance Build (weeks 2–4)**:
- Financial intake script: product-specific for retail vs MSME vs agricultural
- Restructuring option presentation scripts: one per restructuring type per product
- KFS dispatch confirmation and cooling-off management scripts
- NACH amendment advisory script
- Post-restructuring monitoring call scripts (quarterly wellness, re-default protocol)
- Compliance review by lender's legal team: KFS language, cooling-off disclosures, LSP identification, CIBIL advisory accuracy

**Phase 3 — UAT (week 5)**:
- 200+ test scenarios: financial intake variants, all restructuring types, KFS delivery edge cases, cooling-off cancellation, NACH amendment failure handling, multi-product cases, green channel flow
- Credit officer workflow validation: LMS write-back accuracy, case presentation quality, documentation generation

**Phase 4 — Pilot (weeks 6–7)**:
- 500–1,000 restructuring-eligible accounts in 31–60 DPD
- Weekly reporting: intake completion rate, KFS delivery rate, restructuring acceptance rate, cooling-off cancellation rate
- Script and matrix refinement based on pilot outcomes

**Phase 5 — Full Deployment (weeks 8–9)**:
- Full eligible portfolio activated
- Credit officer training on AI-assisted case review dashboard
- Go-live dashboard: daily restructuring volumes, acceptance rates, post-restructuring performance

Joint loan restructuring is operationally more complex than single-borrower restructuring — two people need to be reached, two financial profiles need to be collected, and two sets of consent need to be documented before the credit officer can proceed. Kallix handles this as a sequenced two-call flow.

Joint borrower flow:
1. **Primary borrower call**: AI collects financial intake, presents available restructuring options, and confirms 'I'll also need to speak with your co-borrower [name] to complete this process. Is their mobile number still [registered number]?' If the number has changed, the AI updates it in the LMS (with identity verification of the primary borrower) before calling.
2. **Co-borrower call** (initiated 15–30 minutes after primary borrower call): AI introduces itself: 'I'm calling from [Lender Name] regarding a joint loan account you hold with [Primary Borrower First Name], ending [XXXX]. We are setting up a restructuring arrangement on this account, and I need your consent and a few financial details.'
3. **Co-borrower financial intake**: income, employment status, other obligations — same intake structure as primary borrower. Combined FOIR across both borrowers is calculated.
4. **Consent capture**: both borrowers are asked for explicit verbal consent — 'Do you consent to the restructuring terms I've described?' Double-confirmation recorded.
5. **KFS dispatch**: separate KFS sent to both registered mobile numbers simultaneously — RBI DLG 2022 requires each borrower to receive the disclosure.
6. **Cooling-off**: both cooling-off windows start from the later of the two KFS delivery timestamps, ensuring neither borrower is at a disadvantage.

If the co-borrower is unreachable after 3 attempts over 2 business days, the case is escalated to a human officer who can explore branch-based consent collection or power-of-attorney documentation.

Interest waiver requests are among the most common borrower asks during restructuring discussions — and they require careful handling. The AI's role is to accurately represent what is possible within the lender's policy without making unauthorised promises.

Waiver types and AI handling:

**Penal charges (late payment fees)**: Often waivable at the credit officer's discretion for first-time defaulters with a credible hardship story. Kallix's restructuring matrix allows lenders to configure a 'waiver eligible' flag for penal charges on specific DPD/hardship combinations. When the flag is set, the AI can present: 'Your late charges of Rs 3,450 have been waived as part of this restructuring offer.' This is pre-authorised by the lender's policy, not a discretionary decision.

**Bounce charges**: Rs 250–500 per bounce charge — typically waived in full during restructuring as a goodwill gesture. Configurable in the lender's matrix.

**Interest waiver** (waiving a portion of accrued interest on the outstanding): Requires credit committee approval in most lenders' policies. When the borrower requests this, the AI captures the request verbatim and flags it: 'I've noted your request to reduce the interest component. This requires our credit team's approval. I'll flag this for the credit officer reviewing your case — they'll confirm whether this is possible when they call you within 2 business days.'

**What the AI never does**: promise a waiver that is not pre-authorised in the restructuring matrix. Verbal promises made during a recorded call can be held as binding commitments — a misconfigured or overclaiming AI creates legal liability for the lender.

For cases where the borrower states they will only restructure if a waiver is included, the AI routes to a human negotiator — waiver negotiations require human judgment and credit authority.

Self-employed borrowers represent a structurally different risk profile in restructuring. A salaried borrower who says 'my salary dropped by Rs 20,000' has a predictable new income level. A freelancer, trader, or contractor who says 'my business is slow' has income that could recover in 2 months or remain depressed for 2 years — and an EMI calibrated to the average may still be unaffordable in lean months.

Kallix's self-employed intake adjustments:

**Income capture**: 'Instead of your monthly income, can you tell me what you earned in total over the past 12 months? And approximately how does your income vary — are there months when it's much higher or much lower than average?' This elicits both the annual total and the seasonality pattern.

**Account Aggregator bank statement pull**: For MSME and self-employed borrowers who consent, the AA framework pulls 12 months of bank statements — Kallix identifies the 3 lowest monthly inflows and uses these as the stress-scenario income for EMI calibration. This is more conservative than using averages and more likely to produce a sustainable EMI.

**Seasonal step-up EMI**: For businesses with predictable seasonality (e.g., a retailer with high Q4 income), a step-up EMI structure is presented: lower EMI in lean months (March–July), higher EMI in peak months (October–February). The total annual repayment is the same — the timing is aligned to cash flow.

**ITR reference**: Self-employed borrowers are asked to confirm their last ITR-filed income — this acts as a documented income anchor even if the actual current income is lower. Divergence between ITR income and stated current income is flagged for the credit officer.

**FOIR for self-employed**: calculated on the lean-month income scenario, not the 12-month average — this is Kallix's conservative default. Lenders can configure a blended FOIR (50% lean month + 50% average) if their credit policy permits.

Education loan restructuring has a unique context — these borrowers are typically young (22–28 years), early in their careers, and often dealing with income significantly below their eventual earning potential. The restructuring decision should reflect that a lower EMI now is more sustainable than an OTS that damages their credit history for the most productive years of their career.

Kallix's education loan restructuring framework:

**Post-moratorium repayment phase (most common restructuring context)**:
- The AI asks: 'When did you complete your course, and are you currently employed?'
- If employed, income is captured and FOIR calculated on current salary.
- If unemployed or underemployed: the AI presents a moratorium extension of 6–12 months (if the lender's policy permits for job-seeking borrowers) before restructuring. This is different from a restructuring — the original loan terms remain unchanged, and no 'Restructured' flag is reported.

**CSIS subsidy interaction**: For borrowers from families with income below Rs 4.5 lakh/year, the Central Sector Interest Subsidy Scheme (CSIS) subsidises interest during the moratorium. If the restructuring extends the effective moratorium, CSIS eligibility must be re-verified — the AI notes this and routes to the education loan servicing team for confirmation before extending the moratorium.

**Tenure extension for education loans**: Education loans under IBA Model Scheme can have tenures up to 15 years for loans above Rs 7.5 lakh. If the current repayment schedule is too aggressive, tenure extension to the maximum permissible is the most common restructuring option — it reduces EMI without a 'Restructured' bureau flag if implemented within the original product's maximum tenure.

**Co-applicant (parent/guardian) role**: Most education loans have a parent or guardian as co-applicant. If the student is the primary borrower and is unemployed, the co-applicant's income becomes the primary repayment source during restructuring — the AI captures both income profiles.

Documentation collection is one of the biggest friction points in loan restructuring — borrowers who are already stressed find it difficult to gather and submit physical documents. Kallix's digital-first document collection is designed to remove this barrier entirely.

Digital document collection flow:

**WhatsApp upload**: After the AI intake call, a WhatsApp message is sent with a secure document upload link (hosted on the lender's domain, not a third-party URL). The borrower can photograph and upload documents directly from their phone. OCR validation checks that the uploaded document matches the expected format (e.g., salary slip has employer name, amount, and date visible).

**Account Aggregator consent**: For borrowers with a registered AA account (Finvu, OneMoney, CAMS Finserv, PhonePe AA), a consent request can be sent via SMS — the borrower approves in their AA app and the lender receives a structured bank statement and (for MSME) GST turnover data. This eliminates the need for physical bank statements entirely.

**Minimum documentation for green channel**: For accounts eligible for green channel restructuring (outstanding below Rs 2 lakh), the lender may waive all income documentation — the pre-approved terms are calibrated conservatively enough that full income verification is not required. This is configurable per the lender's credit policy.

**Document validation**: Kallix's OCR layer validates uploaded documents for: (a) legibility; (b) date recency (salary slip must be within 90 days); (c) name match with borrower KYC data. Documents failing validation trigger a re-upload request with specific guidance ('Your salary slip appears to be from July 2024 — please upload a more recent one').

**Credit officer document review**: The credit officer receives a pre-packaged digital case file with all documents validated, income extracted, FOIR calculated, and restructuring recommendation pre-populated — reducing review time from 45–60 minutes to 10–15 minutes per case.

The guarantor dimension of loan restructuring is one of the most legally nuanced aspects — and one that lenders most frequently overlook, with significant consequences. Under Indian contract law (Indian Contract Act 1872, Section 133), a material variation to a loan contract without the guarantor's consent releases the guarantor from their liability. Restructuring — which changes the tenure, EMI amount, or interest terms — is a material variation.

Kallix's guarantor workflow in restructuring:

1. **LMS check**: at intake, the AI's LMS query retrieves whether a guarantor exists on the loan account.
2. **Guarantor notification**: if a guarantor exists, the AI informs the primary borrower: 'Your loan has a guarantor — [Guarantor First Name] — who must also be informed of any change to the loan terms. I'll send them a notification and give them an opportunity to confirm or raise concerns.'
3. **Guarantor notification call**: a separate AI call to the guarantor explaining: (a) the primary borrower is experiencing hardship; (b) the lender is proposing a restructuring; (c) the restructuring will change the guarantor's obligation in the following ways (tenure extension, for example, extends the guarantor's exposure period); (d) the guarantor has 3 days to raise any objections.
4. **Guarantor consent or objection**:
- Consent: recorded in LMS, restructuring proceeds.
- Objection: case escalated to credit officer. Common objections include: guarantor was not aware of the default, guarantor wants to pay off the loan themselves rather than extend, guarantor has their own financial difficulty.
5. **Proceeding without guarantor consent**: legally risky — some lenders include a contractual clause that the guarantor's consent to the original agreement covers reasonable restructuring. Kallix flags this as a legal policy decision for the lender's legal team — the AI does not proceed with restructuring against guarantor objection without explicit lender authorisation.

For MSME loans backed by CGTMSE (credit guarantee), the guarantee fund itself must be notified before restructuring — different from a personal guarantor but equally important.

Misdirecting a NACH-error borrower into a restructuring flow is one of the most harmful miscategorisations in collections — it subjects a paying, non-delinquent borrower to a process that could leave a bureau flag and create unnecessary documentation. Kallix's intake triage prevents this.

Intake triage question: 'Before we discuss options, can you help me understand why the payment wasn't made? Was there an issue with your bank account or mandate, or has your financial situation changed?'

Response categorisation:

**Technical/administrative reason** ('my bank account had an issue', 'I changed my bank account', 'the ECS deduction never came from my bank', 'I paid manually and the NACH also tried to deduct'):
- Route: NACH amendment flow, not restructuring intake
- LMS action: NACH issue flag, no restructuring flag
- AI action: confirms the account is not in financial hardship, assists with mandate fix, and sends a one-time UPI payment link to bring the account current if there is an outstanding bounce

**Duplicate deduction / double debit claim**: AI asks for the deduction dates and amounts from the borrower's bank statement, flags for the lender's transactions team to verify and refund if confirmed.

**Financial hardship** ('I lost my job', 'my salary was cut', 'medical expenses', 'business is slow'):
- Route: full restructuring intake
- LMS action: hardship flag, restructuring intake initiated

**Ambiguous** ('I forgot' / 'I was travelling'):
- Route: PTP capture for immediate payment, not restructuring
- LMS action: standard collection PTP, no restructuring flag

This triage logic saves an estimated 15–20% of restructuring intake calls from becoming unnecessary restructuring cases — reducing the lender's provisioning requirement and protecting borrowers from inadvertent bureau flags.

RBI inspections of bank and NBFC collections and restructuring practices have intensified since 2022, following the RBI DLG Guidelines and several high-profile enforcement actions against lenders for KFS non-compliance and fair practices violations. Kallix's architecture is designed from the ground up for inspection readiness.

Inspection-ready documentation produced by Kallix for each restructured account:

1. **Call recording**: every AI interaction stored as a timestamped audio file in the lender's cloud storage (AWS/Azure/GCP as per lender preference) — retained minimum 90 days, configurable up to 5 years for high-value accounts.
2. **Call transcript**: AI-generated transcript with confidence scoring — flagged sections where the AI's speech recognition had low confidence are marked for human review.
3. **KFS delivery log**: delivery timestamp, channel (WhatsApp/email), delivery status (delivered/read), and document hash (verifying the KFS content was not modified after dispatch).
4. **Cooling-off tracking**: KFS delivery timestamp, 3-day window end timestamp, acceptance timestamp — proving the borrower had the full cooling-off period.
5. **Borrower consent recording**: the specific verbal confirmation of restructuring acceptance, clipped and stored separately for rapid access.
6. **Financial intake data**: structured JSON of all income, expense, and FOIR data collected — shows the basis for the restructuring recommendation.
7. **LMS audit log**: every field change made by the AI integration, with timestamp and API session ID.
8. **Script version record**: which script version was active during each call — with approver and approval date.

For RBI inspectors, the Kallix compliance dashboard provides a single-account drill-down: inspector inputs a loan account number, and all of the above documentation for that account's restructuring interactions is displayed in chronological order. Export as PDF or CSV is available.

Home loans have the longest tenure and largest outstanding of all retail loan products — and their restructuring decisions have the most significant long-term impact on both the borrower and the lender.

Key home loan restructuring considerations:

**Age and tenure**: If a borrower is 50 years old with a 30-year home loan and is seeking a tenure extension from 15 remaining years to 25 years, this pushes the loan maturity to age 75 — beyond standard retirement. Most lenders have a maximum loan maturity age (typically 65–70 years). The AI checks this against the LMS: if the extension would breach the maximum maturity age, it flags for a credit exception review rather than presenting the option as standard.

**Property valuation**: For SARFAESI-eligible home loans at 61+ DPD, the AI mentions that the lender may conduct a property re-valuation as part of the restructuring assessment — this is standard practice and not an indication of imminent enforcement.

**Mortgage release conditions**: The AI clarifies that the mortgage registered with the Sub-Registrar (or the equitable mortgage created by deposit of title deeds) is not released until full loan repayment, even after restructuring. Some borrowers believe a restructuring 'resets' the mortgage — this is incorrect and the AI proactively clarifies it.

**Under-construction property loans**: If the home loan is for an under-construction property and the project is delayed, the AI routes to the home loan servicing team — project delay restructuring involves developer coordination and is outside standard retail restructuring protocols.

**PMAY subsidy interaction**: Home loans with Pradhan Mantri Awas Yojana (PMAY) Credit Linked Subsidy Scheme (CLSS) have subsidy amounts credited to the outstanding at origination. Restructuring must preserve the PMAY subsidy record — the AI flags PMAY-tagged accounts for manual review by the housing finance specialist.

**Joint property / co-ownership**: If the property has multiple owners (not all of whom are loan co-borrowers), the restructuring may need NOC from non-borrower co-owners — flagged for legal team review.

Unauthorised disposal of secured assets is not a collections or restructuring problem — it is a legal problem that changes the entire recovery calculus. A borrower who has sold their mortgaged property cannot be offered standard tenure extension or step-down EMI; they have committed a SARFAESI violation and the lender's legal team must assess the situation before any commercial discussion proceeds.

Common disclosure scenarios:

**Vehicle loan — vehicle sold**: 'I sold the car in March — I needed the money urgently.' This is an RC transfer without NOC from the lender. Under the Hire Purchase Agreement that underpins most vehicle loans, the borrower does not have clear title until the lender's charge is discharged — selling without NOC potentially constitutes fraud. AI response: 'I need to pause our restructuring discussion for a moment. Selling a vehicle that is under a loan without the lender's No Objection Certificate needs to be reviewed by our legal team. They will be in touch within 2 business days. In the meantime, I'll note the details you've shared.'

**Home loan — property sold without NOC**: Even more serious — involves a mortgage registered with the Sub-Registrar. A buyer who purchases a mortgaged property without checking encumbrances acquires an encumbered property. The AI routes immediately to legal.

**Partial asset** (sold one item from a hypothecated stock or equipment): For business loans with floating charge on stock, selling stock in the ordinary course of business is permitted. Selling fixed assets (machinery, equipment) under hypothecation without lender consent is a violation — the distinction matters and the AI routes ambiguous cases to the credit officer.

If the asset disposal is old and the lender was already aware of it (confirmed by LMS flag), the AI treats the account within the lender's already-established legal position — the route has been predetermined by the legal team.

Balance-to-EMI conversion is beneficial to both the borrower (significantly lower effective interest rate) and the lender (converts an uncertain revolving recovery into a structured term repayment). It is the standard restructuring path for credit card accounts in the 31–60 DPD range where the borrower has capacity to repay but cannot manage the full outstanding at once.

Kallix's credit card EMI conversion flow:

**Option presentation**: 'Your credit card outstanding of Rs 68,500 at a revolving rate of 3.4% per month (40.8% APR) can be converted to a 24-month EMI plan at 16% per annum. Your monthly payment would be Rs 3,340 instead of the minimum due of Rs 3,425 — but you'll clear the debt entirely in 24 months rather than revolving indefinitely.'

**APR comparison**: The AI always states both the revolving APR and the EMI APR side by side — RBI DLG 2022 KFS requires the full cost comparison, and this transparency is also the most effective sales tool for the conversion.

**Card status during EMI**: Once balance-to-EMI is accepted, the card's available credit limit is typically frozen or reduced to prevent further spending that would re-inflate the outstanding. The AI explains this: 'During the EMI repayment period, your card's spending limit will be adjusted to [Rs X / frozen] to ensure you can focus on clearing the converted balance.'

**Prepayment option**: The AI mentions that most EMI conversion plans allow prepayment without penalty (subject to lender policy) — a borrower who receives a bonus or windfall can clear the EMI ahead of schedule. This flexibility increases acceptance of the restructuring offer.

**NPA accounts with credit card outstanding**: For 90+ DPD credit card accounts, OTS is the more appropriate path (typically 50–70% of outstanding) rather than EMI conversion — the balance has grown significantly with revolving interest and the borrower's capacity to sustain 24 months of EMI is uncertain. The AI presents OTS for NPA-stage credit cards.

Active litigation between a borrower and lender fundamentally changes the restructuring dynamic — what is said during a restructuring call can be used as evidence in the suit. Kallix's litigation detection is designed to protect both the lender's legal position and the borrower's rights.

Litigation disclosure scenarios and responses:

**Civil suit in district court or DRT**: If the borrower mentions 'I've filed a case in court' or 'my lawyer sent a notice,' the AI responds: 'I've noted that there is legal correspondence between you and [Lender Name]. Our restructuring discussion needs to involve our legal team to ensure any arrangement we make is appropriate given the legal situation. I'll flag this for our legal officer — they'll be in touch within 3 business days.' Call noted as 'borrower in litigation — legal referral.'

**Consumer Forum / NCDRC complaint**: Consumer complaints about unfair practices or mis-selling are increasingly used as a restructuring leverage tactic. The AI acknowledges: 'I'm aware you have a complaint with the consumer forum. Our team will address that separately through the appropriate channel. Regarding your loan outstanding, would you like our legal team to discuss a resolution that covers both the complaint and the payment plan together?'

**Ombudsman complaint**: As covered in the debt recovery module — collection calls suspended, formal grievance channel activated.

**Counter-claim by lender (DRT case pending for large outstanding)**: Some NPA accounts above Rs 20 lakh are already under DRT proceedings by the time the AI interacts with the borrower. If the LMS flags 'DRT case filed,' the AI's role is limited to providing the legal officer's contact details and noting inbound settlement interest — it does not conduct restructuring discussions.

**Out-of-court negotiation during litigation**: In many cases, lenders and borrowers reach an out-of-court settlement even while litigation is ongoing. The AI facilitates the intake for the human legal negotiator who will execute the settlement with proper legal documentation — the AI's role is triage and context collection, not execution.