Indian Consumer Market Research: How to Study Buyers Across India

Before you pay for fieldwork, read the free data. It sizes the market and tells you which quotas to set.

Spending. MoSPI's Household Consumption Expenditure Survey 2023-24 puts average monthly per capita consumption expenditure (MPCE) at ₹4,122 in rural India and ₹6,996 in urban India, excluding items received free through welfare schemes. MoSPI says the urban-rural gap has narrowed to 70%, from 84% in 2011-12 (MoSPI press note).

Spending varies more between states than between rural and urban India as a whole:

MoSPI also reports that non-food items take 53% of average MPCE in rural areas and 60% in urban areas. Within food, "beverages and processed food" is the largest item, followed by milk and milk products.

Internet use. The IAMAI and Kantar "Internet in India 2024" report counts 886 million active internet users in 2024, with 488 million, or 55%, in rural areas. It also says 41% of the population are not active internet users (MediaNama, January 2025). For month-by-month connection data, TRAI publishes performance indicator reports.

Decide your segments before fieldwork, because each one needs enough completes to be read on its own.

• State and region. The HCES table shows how far states differ. A "North India" average can mask a state that behaves differently.
• Urban or rural. Separate them in the sample even if the client reports a single number.
• City tier. Write your definition of Tier 1, 2 and 3, or the list of towns, into the brief so every vendor samples the same places.
• Socio-economic class. NCCS classifies households by ownership of durables and vehicles. In 2024 the Market Research Society of India introduced ISEC, which factors in women's education (Research Live, February 2024). Use the same system across waves so results compare.
• Language. The Eighth Schedule of the Constitution lists 22 languages (background). Field in the language each region reads comfortably, not only in English and Hindi.
• Online or offline. With 41% of people not active online, an online-only study describes connected India. Say so in the report.
• Who decides. The buyer, the user and the person who pays can be three different people. Screen for the role you need.

A national study can combine two: an online sample for the connected majority of buyers in your category, and face-to-face interviews for the groups online panels miss. Report the two separately before you combine them.

Suppose you need a national read on biscuit buyers, with results by zone and by urban or rural. Here is one way to plan 1,200 completes. The structure is the point; your numbers will differ.

This is a disproportionate design: each zone gets the same number of completes, so you can compare zones. Before reporting a national figure, weight each cell back to its share of the population or of category buyers, using Census or HCES figures.

How precise is each cell? These are 95% confidence margins for a result close to 50%, worked out as 1.96 × √(0.25 ÷ n):

Weighting usually widens these margins, so treat them as a floor. If a cell of 150 is too loose for a decision, enlarge that cell rather than the whole sample.

Translation errors are measurement errors. Follow four steps for each language:

1. Translate from the English master into the target language.
2. Have a second translator back-translate it to English without seeing the master.
3. Compare the two and fix any question whose meaning moved.
4. Pilot the translated version with real respondents from that region and read their open-ended answers.

Keep scales consistent. If the English uses a five-point scale, every language version must use five points with equivalent labels.

Understanding a consumer market works in layers, and each layer needs a different tool:

Start at the top and move down only as far as the decision needs. A pricing decision may need only behaviour and a trade-off exercise. A launch into a new category may need all four.

For detailed method guides, see usage and attitude surveys, market segmentation surveys and qualitative research in India.

Your study must represent offline India. We survey through a smartphone app. For the 41% who are not active online, you need face-to-face or phone fieldwork.

You need real purchase data. Surveys capture what people say. For what households actually bought, use a purchase panel; for shop sales, retail measurement.

You need ethnography. In-home visits and diaries need trained field researchers, which a self-serve survey tool does not provide.

Our panel can carry the online cells of the four-zone design, the connected buyers in each zone. The face-to-face cells for offline households need a field agency.

At list price the full 1,200-complete design comes to 1,200 × ₹120 = ₹1,44,000 on Pro (pricing). A single Pro bundle tops out at 1,000, so buy it as two, for example North and South (600) then East and West (600). Translated versions can be piloted first on the Free plan's 30 responses.

For the cells and weights in the plan, the facts that matter are targeting and reach. Quotas can be set by city (any Indian city can be named, region is reached through its cities), city tier, NCCS and age, and the questionnaire can run in as many Indian languages as the study needs. The SuperJ app behind the panel reports a 25M+ consumer footprint, but the members eligible for a given survey are a smaller group, so check rural cells in each zone fill before you commit to the design (panel methodology).