An under-served segment does not announce itself. It shows up in the data as a cluster of signals that individually look like noise but collectively reveal that real demand exists where the institution's products are not reaching. The Product Sales Manager AI reads those signals continuously — from the institution's own pipeline data, from external market signals, and from borrower behaviour patterns — and flags the segment where a product intervention will produce the highest incremental return.
What product-market fit means in a lending context
In a lending portfolio, product-market fit failure has three recognisable signatures. First, the institution receives high volumes of applications from a segment but has low sanction rates — indicating that the product terms do not match what the segment can service. Second, the institution has low application volumes from a segment despite evidence that credit need exists — indicating that the product is not reaching the segment at all. Third, the institution sanctions loans to a segment but has high early-stage NPA — indicating that the product structure does not match the segment's repayment capacity.
Each signature requires a different response. Low sanction rates may indicate that credit policy needs recalibration for the segment's income profile. Low application volumes may indicate that the channel, the product name, or the minimum ticket size is creating a barrier. High early NPA may indicate that the tenure or EMI structure does not align with the segment's cash flow pattern. The Product Sales Manager AI identifies which signature is present, in which segment, and what the response should be.
The 8 product-market fit signals the AI monitors continuously
High application volume, low sanction rate for specific borrower profile
When a specific borrower sub-segment — defined by income type, geography, ticket size, or employment category — shows a sanction rate 20%+ below the portfolio average, the product terms are misaligned with that segment's profile. The gap identifies where the product design, not the credit quality, is creating the friction.
→ Action: Segment-specific product variant with adjusted FOIR ceiling or income computation methodHigh inbound query volume that does not convert to applications
When a segment generates significant inbound queries (call centre, website, DSA enquiries) but a low fraction convert to applications, the product has awareness but a barrier to application. Common causes: minimum ticket size too high, documentation requirement too complex, or interest rate too far from expectations for this segment's reference point.
→ Action: Reduce friction at application entry — smaller minimum ticket, alternative income proof, rate transparencyDPD 0–90 NPA disproportionately concentrated in specific product–segment combination
When early-stage NPAs concentrate in a specific product–segment combination (e.g., 36-month personal loans to Tier 2 self-employed borrowers), the product structure is mismatched to the repayment pattern. A 36-month fixed EMI does not align with a business with quarterly cash flows — even if the total income justifies the loan.
→ Action: Flexible EMI product with step-up structure or quarterly repayment option for this segmentBorrowers applying for a product whose purpose does not match the loan use
When a significant proportion of personal loan applications state "business purposes" as the use, the MSME product is not reaching those borrowers — they are using the path of least resistance instead. This is a product distribution failure, not a credit failure: the segment needs an MSME product but is accessing a personal loan because the MSME pathway is harder to navigate.
→ Action: MSME product simplification or DSA incentive restructure toward business loansDSAs in a geography shifting volume to competitors — detected from volume drop patterns
When DSA-sourced application volume from a specific geography drops more than 25% in a quarter without an identifiable market reason, DSAs are likely re-routing to a competitor with better product terms. This is a product-market fit failure at the DSA level — the product is not competitive enough to retain intermediary loyalty in that geography or segment.
→ Action: DSA conversation + competitor rate intelligence + product position review for geographyGrowing business population in a geography with no increase in MSME loan applications
When GST registration data shows significant new business formation in a geography but MSME loan applications from that geography are not growing proportionally, the product is not reaching the new business population. This is an awareness or access problem — the product exists but the channel has not reached the new segment.
→ Action: New DSA empanelment or digital campaign targeting new GST registrations in the geographySignificant proportion of approvals at the maximum eligible amount — demand exceeds supply
When more than 30% of approved loans are at the maximum eligible amount under the current FOIR ceiling, borrowers are being constrained by the product limit, not by their demand. Raising the effective limit — either by adjusting the FOIR ceiling for strong profiles or by introducing a premium product tier — would convert existing borrowers into larger-ticket customers.
→ Action: Premium product variant with higher FOIR ceiling for CIBIL 750+ profilesApplication volumes spike in specific months but product terms do not accommodate seasonal cash flows
Working capital demand for MSMEs in agriculture-adjacent sectors, festive retail, and construction spikes 3–4× in specific months. A fixed monthly EMI product does not accommodate a borrower who has high cash flow in Q3 and low cash flow in Q1. A bullet or seasonal repayment structure would serve this segment better — and reduce their NPA risk.
→ Action: Seasonal repayment product — higher EMIs in peak months, lower in lean monthsA live signal cluster: the under-served woman entrepreneur segment
From signal to recommendation: what the AI produces
When the signal accumulation threshold is crossed — typically 3 or more signals active simultaneously for the same segment — the Product Sales Manager AI generates a product recommendation brief: the segment definition (who they are), the evidence for under-service (which signals are active, with data), the estimated addressable opportunity (how many eligible borrowers, estimated credit demand), the proposed product intervention (what changes to credit policy, documentation, tenure, or pricing would address the gap), and the business case (incremental NII if the segment is reached at a 10% market share). The recommendation goes to the product committee as a data-backed brief, not a market intuition.
Under-service is not a market failure — it is a product design failure
A segment that is under-served is not under-served because it is unworthy of credit. It is under-served because the institution's product — its documentation requirements, its income assessment model, its minimum ticket size, its EMI structure — was designed for a different segment. The Product Sales Manager AI identifies the specific product design elements that are creating the barrier, not just the existence of the gap. The gap is the diagnosis. The product intervention is the treatment. And the business case is the reason the product committee should prioritise it this quarter rather than next year.