What Is a Survey? Types, Methods & How to Run One in India

About Anaxee

40,000+ Digital Runners · 540 districts · 26 states · 11,000+ pincodes of reach

★ Winner, National Startup Award (Govt. of India)Surveys run across all 26 statesIndia’s Reach Engine since 2016

A survey turns a question you cannot answer from a desk – what do people think, do, or have? – into data you can act on. Businesses use surveys to understand customers and markets; governments and NGOs use them to measure programmes and reach; researchers use them to test ideas. Whatever the purpose, a survey is only as good as the way it is designed and the way the data is actually collected.

This guide explains what a survey is, the main types and methods, how to run one well, and – the part that decides everything in a country like India – how to collect survey data you can trust at scale.

What is a survey?

A survey is a systematic way of collecting information from people by asking them questions, in a consistent format, so the answers can be compared and analysed. The group you ask is usually a sample chosen to represent a larger population, and the questions make up the questionnaire. Done well, a survey lets you learn about thousands or millions of people by carefully asking a representative few.

Types of surveys

Surveys are classified a few different ways – and most real surveys are a combination. For the full breakdown with real examples, see the guide to the types of surveys.

• By purpose. Descriptive (what is happening – market size, coverage), analytical / explanatory (why – what drives a behaviour), or exploratory (early understanding of a new question).
• By timing. Cross-sectional (a snapshot at one point in time) or longitudinal / panel (the same respondents tracked over time, as in baseline-endline studies).
• By coverage. A census (everyone in the population) or a sample survey (a representative subset) – most surveys are sample-based.
• By mode. In-person (CAPI), telephonic (CATI), paper (PAPI) or online – see the guide to choosing a survey mode.
• By subject. Market and consumer research, household surveys, retail and shop surveys, farmer and agri surveys, and programme monitoring and evaluation.

How a survey works, step by step

1. Define the objective. The one decision the survey must inform – this sets every question.
2. Design the questionnaire. Clear, unbiased questions, mostly closed-ended, with skip-logic and validation – see how to design survey questions.
3. Choose the sample. Who to ask and how many, with quotas so the sample represents the population.
4. Collect the data. Run the fieldwork or calls, geo-tagged and time-stamped so every response is traceable.
5. Quality-check. Back-check a sample, flag anomalies, re-verify what does not add up.
6. Analyse. Clean, weight and turn the responses into findings and decisions.

Survey vs questionnaire – what is the difference?

They are often confused. A questionnaire is the set of questions; a survey is the whole exercise – defining the objective, sampling, collecting, verifying and analysing. The questionnaire is one part of the survey, not the survey itself.

Where surveys get hard in India

Designing a good survey is the easy half. The hard half is collecting the data – reaching a representative sample across a country of 650,000+ villages and 20+ languages, mostly offline, and making sure every response was genuinely collected in the field rather than invented at a desk. That is a last-mile and verification problem before it is a research one, and it is exactly where large surveys quietly break.

Where technology and AI change surveys

A survey used to be only as honest as its least-supervised enumerator. Technology changes that. Anaxee runs surveys on an app: every response is geo-tagged, time-stamped and photo-backed, flowing to a live dashboard you can watch fill in by district and quota. AI flags responses collected too fast to be real, duplicates, and locations off the assigned beat – so fabricated or weak data is caught during the survey, not after the report ships. That is the difference between a survey you hope is right and one you can prove is.

How Anaxee helps

Anaxee – India’s Reach Engine, a National Startup Award winner – runs large-scale field surveys and data collection through 40,000+ field executives across 540 districts: in-person (CAPI) and telephonic (CATI) surveys, household, retail, agri and M&E surveys – every response verified, on a live dashboard. Run it fully managed, or on the Anaxee Partner app for your own team.

Proof – surveys run at scale

2,42,672farmers surveyed · 3,667 villagesDeHaatRead case study →463respondents · 26 statesUrjanetRead case study →250+farmers · field researchUniversity of IllinoisRead case study →

Surveys run for research, agri, utility and development clients across all 26 states. See all case studies →

Related reading

Large-scale field surveys →Which survey mode? →Data collection in India →

What happens when you reach out to Anaxee

1. Tell us the survey – the question, the sample and the markets – by form, WhatsApp or a call.
2. A real person calls back – our Indore team, within one working day.
3. We run a pilot – verified survey data from a region, on a dashboard you can open yourself, before you scale.

A 150+ person team in Indore, a telecalling command centre and 40,000+ field executives – a team that runs surveys every day, not an inbox that goes quiet.

Frequently asked questions

The bottom line

A survey is how you turn a question about many people into data you can trust – but only if the sample represents the population and the data was genuinely collected, not invented. Know the types, follow the process, and for surveys at scale in India, use a field network that verifies every response. Define the question; we will bring the answers.