Merchant Awareness and Adoption of Credit Lines on UPI among Small Retailers in Pune (SIP, SPSS)

Merchant Awareness and Adoption of Credit Lines on UPI is a Summer Internship Project (SIP) anchored at a modelled Pune-based merchant-lending services firm (fictional name: Sahyadri Finserve Solutions, stated as fictional throughout the report). The regulatory hook is real: the Reserve Bank of India's directive of 4 September 2023 allowed scheduled commercial banks to let customers pay through pre-sanctioned credit lines over UPI, and NPCI's Credit Line on UPI product made the facility work across payment apps. Small merchants sit on both sides of this facility — they accept UPI hundreds of times a day, yet their own working capital still comes mostly from wholesaler credit (udhaar), trader advances or informal moneylenders.

The study covers roughly 190 small retailers (kirana, pharmacies, hardware, stationery and garments) across four Pune zones and measures: how many have heard of the facility, how many use or intend to use it, what they currently borrow against, and which factors — digital-literacy score, monthly UPI turnover, perceived usefulness, perceived cost and risk — actually drive adoption intention.

Data from a structured bilingual (English/Marathi) questionnaire is analysed in IBM SPSS Statistics (chi-square, one-way ANOVA, Pearson correlation and multiple regression), with MS Excel for cleaning and a small Tableau Public dashboard showing the zone-wise awareness map. The deliverable to the host firm is a merchant-segment awareness and adoption matrix its onboarding team can act on.

Small retailers form the backbone of Pune's neighbourhood commerce, yet their working capital is financed informally: wholesaler credit tied to purchase obligations, advances from traders, or moneylenders charging steep monthly rates. The RBI's 2023 framework for pre-sanctioned credit lines on UPI and NPCI's Credit Line on UPI product offer a formal alternative — a bank-sanctioned line the merchant can draw at the QR code, use only when needed and repay flexibly. But a facility only helps people who know it exists and trust it.

The host firm's onboarding data (modelled) suggests wide, uneven awareness: merchants in the same market, with similar turnover, show completely different borrowing behaviour. No current study segments small merchants by category, zone, digital-literacy and current credit source to explain this gap. Without such evidence, the firm markets uniformly and wastes effort on merchants who cannot adopt, while genuinely constrained merchants stay unserved.

There is therefore a need for a structured survey study that measures awareness, current credit behaviour, perceived usefulness/perceived risk and adoption intention among Pune's small merchants, and statistically identifies which factors predict adoption — so that awareness campaigns and onboarding offers can be targeted segment by segment.

Abstract

This SIP studies whether small merchants in Pune are aware of, and willing to adopt, pre-sanctioned credit lines that can be spent through UPI — a facility the Reserve Bank of India opened in September 2023 and NPCI operationalised as Credit Line on UPI. Using a structured survey of about 190 small retailers across four Pune zones and SPSS-based hypothesis testing, the study identifies who knows about the facility, who uses it, and which measurable factors predict adoption intention.

Introduction

India's merchant digital-payments story is usually told from the payer's side. The credit side is newer: a pre-sanctioned credit line sanctioned by a scheduled commercial bank can now be linked to a UPI ID and spent at any UPI QR code, with the customer's prior consent, repaid later like any credit facility. For a kirana owner this is potentially the first formal, collateral-free, on-demand working-capital line they have ever held — sanctioned once, drawn as needed, interest paid only on what is used.

The SIP is hosted at Sahyadri Finserve Solutions, a modelled merchant-onboarding and lending-support firm in Pune that assists banks in acquiring and servicing small-retailer credit customers. Its onboarding team repeatedly observes two merchants with identical UPI turnover behaving completely differently: one borrows formally, one never does. Understanding why is a commercial question for the firm and a financial-inclusion question in general.

Existing system / literature gap

• Published research on UPI adoption overwhelmingly studies consumers (adoption intention, trust, perceived risk) rather than merchants as borrowers.
• Studies on merchant finance (MUDRA loans, credit cards for MSMEs) predate the 2023 credit-line-on-UPI framework, so there is little survey evidence on awareness of this specific facility at the counter level.
• Industry reports describe aggregate UPI volumes but do not segment by shop category, zone or digital-literacy level — exactly the segmentation a lending-support firm needs.

The proposed study

A descriptive, cross-sectional survey of ~190 small retailers (kirana, pharmacy, hardware, stationery, garments) sampled from four Pune zones. The questionnaire has five sections: business profile, current credit sources, UPI usage maturity, awareness of Credit Line on UPI, and Likert-scale items on perceived usefulness, ease, cost and risk feeding an adoption-intention score. Analysis in SPSS: descriptives, chi-square (category × awareness), ANOVA (digital-literacy score across age groups), Pearson correlation (UPI turnover × attitude) and multiple regression (adoption-intention score). Findings feed a segment-wise recommendation matrix for the host firm.

Feasibility

• Technical: Excel and SPSS are covered in the SPPU curriculum (GE-IL-04, GE-IL-11); Tableau Public is free. No special software is needed.
• Economic: Data collection is on foot within Pune; cost is limited to travel and printing (under ₹1,500).
• Operational: Merchants answer in 10–12 minutes in Marathi or English during low-traffic hours; the questionnaire collects no names, phone numbers or government ID numbers, so consent is easy to obtain.

In scope

• Small retailers (kirana, pharmacy, hardware, stationery, garments) with a visible UPI QR code, in four Pune zones: central Peths, Kothrud–Warje, Hadapsar–Magarpatta fringe and Pimpri-Chinchwad.
• Awareness of and intention towards pre-sanctioned Credit Lines on UPI sanctioned by scheduled commercial banks (the RBI 2023 framework).
• Current informal/formal borrowing behaviour as reported by the merchant.
• Statistical analysis in SPSS of one wave of survey data (~190 responses) collected over 8 weeks.

Out of scope

• NBFC-issued credit lines on UPI (the RBI initially enabled banks; later extensions are noted in the literature review only).
• Consumers/home buyers; e-commerce sellers without a physical counter.
• Actual transaction data or repayment data of any lender — no internal bank data is used.
• Longitudinal measurement of real adoption after onboarding (suggested as future work).
• Any collection of merchant names, phone numbers, Aadhaar, PAN or GST numbers.

Research design: descriptive, cross-sectional survey design with hypothesis testing. The unit of study is the individual retail outlet; the unit of analysis is the merchant's questionnaire record.

Sampling: stratified convenience sampling. Four zones form the strata; within each zone, markets are walked and every nth shop with a visible UPI QR code is approached until the zone quota is met (target ~48 per zone, n ≈ 190). Merchants under 18 are excluded; one response per shop.

Instrument: a structured questionnaire in English and Marathi — Section A business profile (category, years in business, employees); Section B current credit sources and typical monthly purchase-credit need; Section C UPI usage (years on UPI, approximate monthly UPI receipts, apps used); Section D awareness and usage of Credit Line on UPI (aided recall with a one-page explainer card); Section E 12 Likert items (1–5) on perceived usefulness, ease of use, cost, risk/trust and adoption intention. A 5-item digital-literacy checklist (UPI PIN change, QR payment, statement download, refund awareness, bill-payment apps) is embedded.

Pilot: 15 merchants in a non-sample market; Cronbach's alpha computed in SPSS for each multi-item scale (target ≥ 0.7); ambiguous Marathi wordings revised.

Hypotheses: H1: business category and awareness are associated (chi-square). H2: mean digital-literacy score differs across age groups (ANOVA). H3: monthly UPI turnover correlates positively with attitude score (Pearson). H4: adoption-intention score is predicted by awareness, digital literacy, perceived usefulness, perceived risk and credit constraint (multiple regression, α = 0.05).

Ethics: verbal informed consent; no personal identifiers; merchants may skip questions; data stored in a password-protected file and reported only in aggregate.

The study is designed as a single pipeline from regulatory context to a targeting matrix. Two data streams — the merchant survey and the firm's modelled onboarding observations — are analysed separately in SPSS and Excel, then merged conceptually at the interpretation stage.

flowchart TD
  A["RBI 2023 directive + NPCI Credit Line on UPI"] --> B["Literature review: UPI adoption, merchant finance"]
  B --> C["Objectives and 4 hypotheses with guide"]
  C --> D["Bilingual questionnaire: 5 sections, 12 Likert items"]
  D --> E["Pilot 15 merchants; Cronbach alpha >= 0.7"]
  E --> F["Stratified sample ~190 merchants across 4 Pune zones"]
  F --> G["Excel: coding, cleaning, composite scores"]
  G --> H["SPSS: chi-square, ANOVA, correlation, regression"]
  H --> I["Hypothesis verdicts"]
  I --> J["Awareness-adoption matrix by category and zone"]
  I --> K["Regression: drivers of adoption intention"]
  J --> L["Tableau zone dashboard"]
  K --> L
  L --> M["Segment-wise recommendations to host firm"]

Variable framework

flowchart LR
  V1["Business category"] --> O
  V2["Digital-literacy score"] --> O["Adoption-intention score"]
  V3["Monthly UPI turnover"] --> O
  V4["Perceived usefulness / risk"] --> O
  V5["Current-credit constraint"] --> O
  O --> R["Segment matrix for onboarding"]

Measures. Awareness is binary (aided recall) plus a source-of-information check. Composite scores are means of their Likert items after reverse-coding negative statements. The regression model is checked for multicollinearity (VIF < 5) and the normality of residuals before interpretation, so every claim in the findings chapter can be traced to a test the examiner can rerun.

• Post-onboarding tracking: a follow-up wave six months later to measure actual activation of credit lines, turning the intention model into a behaviour model.
• NBFC extension: once credit lines from NBFCs flow through UPI, repeat the study with NBFC-sourced lines and compare sanction speed and limits.
• City comparison: replicate in Nashik and Nagpur to see whether tier effects dominate zone effects.
• Logistic classification: replace the linear intention model with logistic regression predicting actual uptake, usable directly by the firm's CRM.
• Qualitative layer: 10 merchant interviews on trust and repayment anxieties to explain the regression residuals the survey cannot.