Cost drivers, quote lines and the questions to ask before you sign
If you are asking how much does a survey cost in India, the honest answer is that no single rate card exists. The budget comes from a handful of drivers: completed interviews, interview length, travel between sampling points, language, mode and how much verification you want. This guide shows each driver, how to read a quote and how to cut cost without cutting quality.
A survey budget in India is mostly people-time. Cost rises with the number of completed interviews, the minutes each takes, the distance field teams travel between clusters, the languages involved and the share of interviews that supervisors re-check. Design, programming, analysis and reporting add a fixed layer on top of the variable field cost.
Think of it as two budgets. The fixed part covers questionnaire design, translation, app programming, training and the report; it barely changes whether you interview 800 or 3,000 people. The variable part is field time: enumerator days, supervisor days, travel, stay and respondent contact attempts. Large samples spread the fixed part thin, which is why the cost per interview usually falls as the sample grows.
What does not show on most quotes is the cost of failure: interviews that must be redone because GPS shows the enumerator never reached the village, or a questionnaire that confused respondents and had to be re-fielded. Paying for checks up front is usually cheaper than paying for a second round.
Each card is a lever. Change one and the quote moves; ask a vendor which lever explains any line you do not understand.
The sample size, counted as finished and accepted interviews, not attempts. Rural door-to-door work may need two or three visits to finish one interview.
A 12-minute interview and a 45-minute interview are different products. Longer interviews cut the number an enumerator can finish each day, and fatigue lowers answer quality.
Ten interviews in each of 200 villages costs far more than 50 in each of 40, because travel days replace interview days. Clustering saves money but widens the margin of error.
Each language needs translation, back-translation and interviewers who speak it. Doctors, business owners and officials take longer to reach than households.
Face-to-face with an app, telephone and online panels have very different cost structures. Mode also changes who you can reach, so price alone should not decide it.
GPS and time stamps, audio consent, photo evidence and supervisor revisits all add cost. They are also what lets you trust the data in front of a funder or a board.
Clean data with a codebook is one deliverable. Weighted tables, district-level estimates, dashboards and a written report with recommendations are several more.
Market research cost in India varies because a respondent is not a fixed unit. A short intercept interview in a city market, a 40-minute household interview in a remote block and a scheduled interview with a hospital buyer can differ several times over in field effort, even though each counts as one respondent.
A single per-respondent figure bundles incidence (how many people you screen to find one eligible respondent), interview length, travel and checks into one number. When incidence is low, say owners of a specific appliance brand, most field time goes into screening people who do not qualify. Ask the vendor to state the incidence they assumed; if the real figure is lower, the cost or the timeline will change.
Consumer studies in towns usually have high incidence and short travel. B2B studies need appointments, senior interviewers and repeated follow-ups. Studies of migrant workers, tribal households or patients need careful consent and often a local guide. Each category should be costed separately rather than averaged into one rate.
Put quotes from different agencies into the same table before comparing totals. A lower total often means a missing line, not a cheaper study.
| Quote line | What it should state | Red flag |
|---|---|---|
| Design and questionnaire | Number of drafts, who signs off, pilot included | No pilot mentioned |
| Translation | Languages, back-translation, who checks | "Translation included" with no process |
| Programming | App or platform, skip logic, range checks | Paper forms with later data entry, unpriced |
| Training | Days, trainer, mock interviews, test | Half-day briefing only |
| Fieldwork | Completed interviews, clusters, days, supervisors | Price per attempt rather than per complete |
| Quality control | Back-check share, GPS, audio, rejection rule | No rejection or replacement rule |
| Data and analysis | Clean data, codebook, tables, weighting | Raw export only |
| Report and presentation | Pages or format, revision rounds | Unlimited revisions promised at no cost |
| Taxes and payment terms | GST treatment, milestones | Full advance before fieldwork plan |
Use the same sample size, interview length and deliverables in every request so the totals are comparable.
Door to door survey cost per household is driven less by the interview itself than by getting to the door. Travel between villages, finding the sampled house, revisiting when nobody is home and supervisor back-checks can take more field time than the conversation, especially in hilly, flooded or sparsely populated areas.
Ask the vendor to walk you through one enumerator day. It shows whether the budget is realistic.
Enumerators in dense urban wards may finish many more interviews a day than colleagues in scattered hamlets, and monsoon months cut field days further. Local-language needs, security clearances and gated housing societies in cities also add time. A vendor who quotes one national rate for every district is either averaging heavily or has not planned the route.
Most savings come from design choices made before fieldwork, not from bargaining over the per-interview rate.
How Bharat Survey scopes software seats and field projects.
Learn moreWhat changes the budget for a constituency or state survey.
Learn moreQuestions to ask before you shortlist.
Learn moreHow mode affects cost and coverage.
Learn moreShorter, clearer questionnaires cost less to field.
Learn moreHow government buyers evaluate cost and quality.
Learn moreStraight answers to the questions people ask most about this topic.
A national or multi-state survey is priced from the same drivers as a small one, but travel, languages, supervision layers and data management grow quickly. Budget separately for design, each state's fieldwork and central quality control. Ask for a state-wise breakdown so you can drop or phase regions if the total is too high.
Only when every quote assumes the same interview length, incidence, geography and checks. Otherwise a low per-respondent rate may hide fewer back-checks, attempts counted as completes or missing analysis. Compare line by line using the same scope, then look at the per-respondent figure.
Online surveys remove travel and enumerator time, so field cost is lower. They also reach a narrower population: people who are online, literate in the survey language and willing to join panels. For rural households, older people or low-income groups, a cheaper online sample may not represent the people you need.
Central government tenders and award details are published on the Central Public Procurement Portal and GeM. Reading awarded contracts for similar studies gives a realistic public benchmark. Check scope, sample and duration before comparing, because tender values include deliverables that vary widely.
Software is priced by seats and field projects are priced on quotation after scoping. We ask for the respondent group, geography, sample, interview length, languages and deliverables, then send a line-by-line proposal showing the assumptions behind each figure.
Share the respondents, geography, sample and deliverables. We will send a line-by-line proposal with every assumption written down.
