# Ethnographic Research in India: Observing Consumers at Home, in Shops and on Their Phones

Ethnographic research in India: in-home immersions, shop-alongs, mobile ethnography, how many households, consent rules, coding and the follow-up survey.

Published 2026-11-02 · Updated 2026-11-02
Canonical: https://hercules.works/blogs/ethnography-research-india

Ethnographic research observes people in their own setting, at home, in a shop or through their phone. In India it needs small samples, fieldworkers who speak the household's language, consent from everyone on camera and a schedule that respects festivals and the monsoon. It explains what people do and why, but cannot say how common a behaviour is; a survey does that. Hercules Works runs that follow-up survey among verified Indian consumers and can recruit for qualitative work through Enterprise. We do not conduct ethnography ourselves.

## Watching Instead of Asking

An interview asks a person to remember and explain. Ethnographic market research watches what they do and asks about it on the spot. That gap is where teams find the detergent decanted into an old tub or the second phone used only for payments. Nobody reports those things in a survey because nobody thinks they matter.

If you are an insights lead planning your first round of fieldwork, the hard questions are how many households, how long, and what to do with the footage afterwards. This page sits beside our guides on [qualitative research](/blogs/qualitative-research-india), [diary studies](/blogs/diary-study-india) and [mixed methods](/blogs/mixed-methods-research-india).

## What Ethnography Is and Is Not

Ethnography comes from anthropology. The Open Encyclopedia of Anthropology defines it as "the recording and analysis of a culture or society, usually based on participant-observation". The anthropologist lives long-term among the people studied and "will leave after one or two years" ([Howell, Open Encyclopedia of Anthropology, 2018](https://anthroencyclopedia.com/entry/ethnography)).

Commercial ethnography is shorter. David Gray's methods textbook says ethnography "typically employs three kinds of data collection methods: observation, interviews and documents". Modern studies, it adds, may visit a site for a two-week period every few months rather than stay for six months to two years ([Gray, Doing Research in the Real World, 3rd ed.](https://sfu.ca/~palys/Gray-DoingResearchInTheRealWorld-Ethnography.pdf)).

So be precise about what you are buying. A single home visit with a discussion guide is an in-home interview. One trip to a shop is a shop-along. Call it ethnography only when you observe behaviour in its own setting, over enough time for the routine to show, and analyse what it means to the people doing it. Nielsen Norman Group treats ethnography as sustained, near-total immersion, and direct observation and contextual inquiry as lighter field methods ([NN/g, Field Studies, 2024](https://www.nngroup.com/articles/field-studies/)).

## Four Ways to Observe: Immersion, Shop-Along, Mobile, Diary

**In-home immersion.** You spend hours in a household watching a routine happen: the morning tea, the weekly wash, the evening screen time. You ask as things happen, not afterwards.

**Shop-along.** An interviewer "accompanies a consenting respondent as they go shopping in order to observe shopping behavior" ([Quirk's glossary](https://www.quirks.com/glossary/accompanied-shopping)). In India that means kirana counters and weekly markets as much as modern trade. Watch what the shopper asks for by name and what the shopkeeper hands over instead.

**Mobile ethnography.** Participants record their own lives. Quirk's defines it as a method that "uses mobile devices to capture in-the-moment insights from participants as they go about their daily lives", through photos, videos and text entries ([Quirk's glossary](https://www.quirks.com/glossary/mobile-ethnography)). It reaches Tier 2 and Tier 3 towns without travel, and misses whatever the participant chooses not to film.

**Diary-based observation.** A diary asks for the same short entry repeatedly over days or weeks. It is self-report rather than observation, but it records sequence in a way one visit cannot. Pair a diary with one or two home visits and you have a light ethnography; our [diary study guide](/blogs/diary-study-india) covers design and drop-off.

## Decision Table: Ethnography, Interview, Diary or Survey

| Your question | Best method | Why it fits | Weak spot |
|---|---|---|---|
| What do people actually do, step by step? | In-home immersion | Behaviour in its setting, probed as it happens | Few households; people act differently when watched |
| Why did one person choose or reject it? | In-depth interview | A full personal account in one sitting | Recall and after-the-fact reasoning |
| What happens across a week, in what order? | Diary or mobile ethnography | Entries close to the moment | Self-chosen moments; drop-off |
| How is the choice made at the counter? | Shop-along | Sees substitution, the shopkeeper's role, the price check | One trip per person |
| How many people do this, in which segments? | Quantitative survey | Sized sample, fixed questions | Cannot find a behaviour you did not ask about |

If you cannot yet name the behaviour, start with observation. If you can name it and want to count it, go to the survey. Our [qualitative research guide](/blogs/qualitative-research-india) covers interviews and groups, and the [mixed methods guide](/blogs/mixed-methods-research-india) covers how the strands connect.

## How Many Households, How Long, and the Cost in Time

We have not found a published Indian norm for household counts, nor an ESOMAR rule. A 2022 systematic review found that studies with empirical data reached saturation "within a narrow range of interviews (9-17) or focus group discussions (4-8)". That held mainly for homogeneous populations and narrowly defined objectives ([Hennink and Kaiser, Social Science & Medicine, 2022](https://pubmed.ncbi.nlm.nih.gov/34785096/)). Treat that range as an upper bound per distinct group, not a target.

Practitioners describe smaller samples. One provider says ethnographic samples are "commonly less than 10" ([FlexMR](https://blog.flexmr.net/mobile-ethnography-market-research)). Another describes in-home work as "almost always limited to 8 to 12 participants, one market, and several weeks of fieldwork". It puts mobile studies at 50 or more participants across several markets, run for at least a week ([Indeemo, March 2026](https://indeemo.com/resources/blog/consumer-research-depth-and-scale)). These are the vendors' own figures; we have not audited them. Count groups, not people: two cities, two life stages and two user types make eight cells, and one household per cell shows what is possible, not what is usual.

Budget in time, not rupees; we have found no published Indian rate card for ethnography to cite. Time runs in four blocks. Recruitment comes first, and it is slower for lapsed users and Tier 3 households. Fieldwork follows the textbook pattern of repeated visits rather than one long stay ([Gray](https://sfu.ca/~palys/Gray-DoingResearchInTheRealWorld-Ethnography.pdf)). Analysis is the block teams under-plan, because video and multilingual notes take long to review. The sizing survey below is the fourth.

## Fieldwork in India: Languages, Small Towns, Consent and Seasons

**Languages.** The 2011 Census counted 121 languages and 270 mother tongues with 10,000 or more speakers each, rationalised from 19,569 raw returns. In the same count, 96.71% of the population named one of the 22 Scheduled Languages as their mother tongue ([Census of India 2011, Language, Paper 1 of 2018](https://cdn.downtoearth.org.in/library/0.58521300_1530794876_census-of-india-2011.pdf)). The moderator should speak the household's language, not work through a translator. Write field notes in that language and translate quotes at the end.

**Tier 2 and Tier 3 access.** Recruit through local contacts and allow travel days. Expect the whole household, and sometimes neighbours, to join a visit; decide in advance whether that is data or noise.

**Consent and recording.** The ICC/ESOMAR Code requires voluntary participation based on clear information about the research, given at the beginning, with the right to withdraw at any time. It lists "picture, sound or video recording" among direct identifiers ([ICC/ESOMAR International Code, 2025 revision](https://community.esomar.org/uploads/public/knowledge-and-standards/codes-and-guidelines/ICCESOMAR-International-Code_English.pdf)). The ESOMAR/GRBN primary data collection guideline names ethnographic studies in its scope. Its mystery shopping section adds that "no personal data of any kind (including photos or recordings)" may be collected about people who have not consented. That covers other shoppers and store personnel ([ESOMAR/GRBN, 2021](https://commbox.org/dl/68e24e6085303c7d1c7e8d7c)). Apply the same rule to a shop-along. In a home, get consent from each adult on camera and plan for children.

India's Digital Personal Data Protection Act, 2023 requires consent that is "free, specific, informed, unconditional and unambiguous with a clear affirmative action". It also requires a notice naming the data and purpose before the request, and withdrawal as easy as consent. It received assent on 11 August 2023 and comes into force on dates the Central Government notifies ([Digital Personal Data Protection Act, 2023](https://www.trai.gov.in/sites/default/files/2024-09/DPDT_11082024.pdf)).

**Festivals and seasons.** Cooking, shopping and spending shift around Navratri, Dussehra and Diwali, and around Ramzan, Onam, Pongal and Durga Puja by region. The monsoon changes commuting, laundry and what is in the market. Avoid festival weeks for a baseline routine, field inside them for festive buying, and say which you did. Festival dates move each year, so check that year's calendar.

## Template: An Observation Guide for a Home Visit

An observation guide lists what to look at, what to ask only when it does not interrupt, and what to record, one per visit.

| Block | What to observe | Ask only if natural | Record |
|---|---|---|---|
| Setting | Where the product is kept, what sits next to it, who can reach it | Who decided where it lives? | Photo of the spot, sketch of the room |
| Sequence | Every step in order, including workarounds | Why that step? (ask after, not during) | Numbered steps with timestamps |
| People | Who does it, who helps, who watches, who overrides | Is it always you? | Roles, not names |
| Substitutes | Other brands, home remedies, loose alternatives in view | When do you use that instead? | Pack photos, with permission |
| Language | Words used for the product, the problem and the result | What do you call this? | Verbatim, in the language spoken |
| Friction | Pauses, repeats, spills, searching, phone checks | What was that? | Timestamp, plus your reading marked as yours |

Keep what you saw separate from what you think it means. Gray's textbook notes that field notes are written "more or less contemporaneously with the events" they describe and are always selective ([Gray](https://sfu.ca/~palys/Gray-DoingResearchInTheRealWorld-Ethnography.pdf)). Write up the same evening.

**What we would do.** We would field two cells before the rest, one in a metro and one in a small town, and write both up before booking the remaining visits. If the guide above survives those two households without a rewrite, you have a guide. If it does not, you have saved yourself the same rewrite across every later visit.

## Analysis: Coding, Themes, Journey Maps, Then the Survey

Code in the spoken language where you can. Braun and Clarke's reflexive thematic analysis names six phases: familiarising yourself with the dataset, coding, generating initial themes, developing and reviewing themes, refining, defining and naming themes, and writing up. The phases "do not prescribe a rigid process you must follow" ([thematicanalysis.net](https://www.thematicanalysis.net/doing-reflexive-ta/)). Code behaviours, not only statements. "Decants into a reused container" is a code; "says the pack is inconvenient" is another. Where they contradict each other is a finding.

Then draw the journey. Nielsen Norman Group defines a journey map as "a visualization of the process that a person goes through in order to accomplish a goal". It lists five components: actor, scenario and expectations, journey phases, actions, mindsets and emotions, and opportunities ([NN/g, Journey Mapping 101](https://www.nngroup.com/articles/journey-mapping-101/)). Build one per segment, not one for the market. Against each step, write how many households you saw do it, out of how many you visited; that describes the people you met, not the city.

Ethnography tells you what to count; the survey counts it. Turn each behaviour code into a closed question with the households' own words as answer options, and add a screener so the sample matches the segment you observed. Our platform fields that survey among verified Indian consumers in the languages the study needs, and codes open-ended answers so you can check the survey's vocabulary against what people said. The [usage and attitude guide](/blogs/usage-attitude-survey-india) covers the questionnaire, the [jobs-to-be-done guide](/blogs/jobs-to-be-done-survey-india) turns workarounds into job statements, and [moderated web call interviews](/use-cases/moderated-web-call-interviews) covers video interviews before the survey.

To be clear: we do not run ethnography. Our panel is app-based, and our research team can recruit qualitative participants through Enterprise plans. The fieldwork itself needs a qualitative agency or your own team.

## Running This Study on Hercules

If your job is the quantitative follow-up to an ethnography, sizing the behaviours you observed among Indian consumers, this is the pilot and main study at list price, with respondents included.

| Step | Plan | Responses | Price per response | Cost |
|---|---|---|---|---|
| Test the questionnaire | Free | 30 | ₹0 | ₹0 |
| Small pilot | Plus | 100 | ₹150 | ₹15,000 |
| Main study | Pro | 400 | ₹120 | ₹48,000 |

Pro is sold in bundles of 400 to 1,000 responses and costs 20% less per response than Plus. Enterprise plans come with a dedicated research team and are priced on request; they also cover recruitment for qualitative work.

The SuperJ network has 2 crore Indian respondents. The survey-eligible panel is a subset, and availability depends on your targeting. Members complete Aadhaar-based KYC once, and each study receives a zero-knowledge proof of verification. Surveys run in as many Indian languages as the study needs. Quality checks flag speeders, straight-lining and inconsistent answers. Read [/panel-methodology](/panel-methodology) and [pricing](/pricing).

The limit: the app reaches smartphone users only, so offline households need face-to-face fieldwork.

## Frequently asked questions

### What is ethnography research in marketing?

Observing consumers in their own setting, at home, in a shop or through their phone. The observation runs over enough time for the routine to show, and you analyse what the behaviour means to them. An interview asks people to recall; ethnography watches.

### What is mobile ethnography?

Participants use their own smartphones to record moments from daily life as photos, video and short text entries over days or weeks. It reaches small towns without travel, but captures only the moments the participant chooses to record.

### How many households does an in-home immersion study need?

There is no published norm. Interview research reaches saturation within about 9 to 17 interviews per similar group, and practitioners describe in-home samples of under 10 to about 12 households in one market. Plan more than one household per distinct group.

### Do I need consent to film inside a home in India?

Yes. The ICC/ESOMAR Code treats picture, sound and video recordings as direct identifiers, and ESOMAR/GRBN guidance bars recording anyone who has not consented. Under the Digital Personal Data Protection Act, consent must be free, specific, informed, unconditional and unambiguous, and it must follow a notice.

### Can a survey replace ethnographic research in India?

No. A survey can only ask about behaviours you already know to ask about. Use ethnography to find the behaviour, then a survey to measure how common it is.