10-Minute Delivery vs the Neighbourhood Kirana: Grocery Channel Choice of Thane Households

Quick-commerce apps promise groceries at the door in 10–20 minutes, and in cities like Thane they have spread from one or two dark stores to almost every residential pocket. Yet the neighbourhood kirana has not disappeared: it still gives credit on the monthly khata, delivers by phone, knows the family's brands and stays open when the app shows "no slots". This research project studies how Thane households divide their grocery spending between quick-commerce apps and kirana stores, and which factors drive that choice.

Primary data is collected from 200 households across four localities — Ghodbunder Road, Naupada, Wagle Estate and Kalwa — chosen to cover high-rise, older middle-class and working-class neighbourhoods. The bilingual questionnaire measures channel usage, share of monthly grocery spend, and five perception constructs on 5-point Likert scales: convenience, price perception, product availability, trust/relationship and credit facility.

The analysis in Excel and SPSS uses chi-square to test association between demographics and the preferred channel, correlation between perceived convenience and usage frequency, one-way ANOVA across income groups and a multiple regression predicting the share of wallet given to apps. The final output is an 80–100-page Mumbai University BMS Semester 6 black book with findings useful for both kirana owners and app marketers.

Abstract

Quick-commerce platforms have changed how urban Indian households buy groceries, but most families still use neighbourhood kirana stores for part of their needs. This study surveys 200 households in Thane to measure the share of grocery spending on each channel and to identify the factors behind channel choice. Using Likert-scale constructs and statistical tests (chi-square, correlation, ANOVA and multiple regression), it examines the role of convenience, price perception, availability, trust and credit. The findings offer practical strategies for kirana owners to retain customers and for platforms to improve basket size.

Introduction

Grocery is the largest and most frequent household purchase category in India. For decades it was served almost entirely by unorganised kirana stores. Supermarkets and e-grocery added options, and after 2021 quick-commerce — dark-store-based delivery in minutes — grew rapidly in metro and satellite cities. Thane, with its mix of new townships and older neighbourhoods, is an ideal setting to observe how different households respond.

Literature gap

Existing studies mostly compare supermarkets with kirana stores or look at online grocery in general. Research specific to quick-commerce is recent, largely industry reports rather than academic surveys, and rarely measures the split of spending instead of a single "preferred" channel. The role of informal credit (khata) and personal relationship, which is central to kirana loyalty, is seldom tested statistically. This study addresses that gap with a household-level survey in a single city.

Proposed study

• Measure channel usage frequency and percentage of monthly grocery spend on apps vs kirana.
• Test whether age, income and household type are associated with the preferred channel.
• Identify which perception factors predict the share of spending on quick-commerce.
• Recommend strategies for both kirana owners and app marketers.

Feasibility

• Technical: questionnaire in Google Forms and on paper; analysis in Excel and SPSS available in the college lab.
• Economic: local travel and printing costs under ₹2,000.
• Operational: respondents approached with the permission of housing-society secretaries and at kirana counters during evening hours; the questionnaire takes about 10 minutes and is available in Marathi for older respondents.

Neighbourhood kirana stores in Thane report fewer small top-up purchases since quick-commerce dark stores opened nearby, while app platforms struggle to win the large monthly staples basket that households still buy on credit from their regular shopkeeper. Both sides make decisions — discounts, delivery radius, credit terms, assortment — with little systematic evidence about why a household chooses one channel for some purchases and the other for the rest.

Popular opinion holds that younger, higher-income families have "shifted to apps", but spending is rarely all-or-nothing. It is unclear how much of the basket actually moves, and whether convenience, price, availability, trust or credit matters most. Without this understanding, kirana owners may compete on price when they should compete on service, and platforms may push discounts when availability is the real issue.

The problem addressed by this study is therefore: to identify the factors that determine the share of grocery spending that Thane households allocate to quick-commerce apps versus neighbourhood kirana stores, and how these factors differ across demographic groups.

Area: Thane city (Thane Municipal Corporation limits) — four localities representing different housing and income profiles.

Unit of study: the household member who does most of the grocery purchasing, aged 18 or above, in a household that has bought groceries from both a kirana store and a quick-commerce app at least once in the last three months.

Content: grocery purchases only (staples, dairy, fruits and vegetables, packaged foods, personal care). Restaurant food delivery, pharmacy and electronics are excluded.

Limitations: non-probability sampling inside each locality, self-reported spending percentages, and one city with fast-changing platform offers — findings describe the situation during the survey period only.

Research design

Descriptive and causal-comparative survey design using primary data, supported by secondary data from industry reports, company press releases and news coverage of quick-commerce.

Sampling

• Population: grocery-purchasing households in Thane city (several lakh households; exact frame unavailable).
• Sampling method: stratified convenience sampling — four localities as strata with 50 households each; within each locality respondents are approached at housing societies (with the secretary's permission) and outside kirana stores in the evening.
• Sample-size logic: for a large population, Cochran's formula n = z²pq/e² with z = 1.96, p = 0.5 and e = 0.07 gives (3.8416 × 0.25) / 0.0049 = 196, rounded to 200. Yamane's formula n = N / (1 + Ne²) with N = 4,00,000 households and e = 0.07 gives ≈ 204, confirming the same order of size. About 240 questionnaires are distributed to allow for incomplete forms.

Questionnaire design

The questionnaire is translated into Marathi and back-translated to check meaning. A pilot with 25 households checks clarity; Cronbach's alpha ≥ 0.70 is required for each construct.

Hypotheses and tests

Significance level α = 0.05.

Timeline (12 weeks)

Weeks 1–2 topic, literature and synopsis approval · Week 3 questionnaire, translation and pilot · Weeks 4–6 fieldwork · Week 7 data entry and cleaning · Weeks 8–9 analysis · Weeks 10–11 writing · Week 12 black-book printing, PPT and viva preparation.