State government record
Recall and use of state schemes, law and order, prices and jobs, measured with specific questions instead of a single approval score.
Election surveys · State assemblies
Assembly seat research where the state government, the MLA and the street all get judged at once
Bharat Survey is a Vidhan Sabha election survey agency for MLA candidates, party state units, newsrooms and researchers. An assembly vote mixes three judgements: the state government's record, the chief ministerial choice and the local MLA's work. We sample each seat by polling-station cluster, interview with GPS-verified field teams, and separate those three judgements so the campaign knows which one is moving votes.
- Seat samples built from polling-station clusters
- State, leadership and MLA questions asked separately
- Many seats on one method and one dashboard
- Margins and unweighted bases on every table
3,000
Target homes
2,184
Visits
1,962
Completed
1,871
Accepted
Daily accepted surveysLast 12 days
Quality
Accepted1871
In review64
Outside area19
Weak signal8
Top issues · by panchayat
Water62%
Roads48%
Power31%
Drains24%
Home #17 · Visit outcome
- Complete
- Partial
- Declined
- No contact
- Home locked
- Revisit
Assembly seat samplingMLA report cardsChief minister preferenceState scheme recallAnti-incumbency checksMulti-seat trackingExit poll embargoNeutral method
Why an assembly seat needs its own design
One preference question hides why people vote. We split it into the judgements voters actually make.
Leadership choice
Preferred chief minister asked on its own, with names rotated, so a popular leader is not confused with a popular local candidate.
MLA performance
Awareness of the sitting MLA, visits to the area and work on roads, water and schools, kept apart from party preference.
Local issues
Issue lists written per seat. A canal-irrigated rural seat and a city seat do not share a questionnaire.
Fieldwork you can watch
During a pre-poll round the campaign or newsroom sees interviews arrive on a map as phones sync. Each point is an interview with a time, an interviewer and a location check against the assigned polling-station cluster.
Points outside the cluster, very short interviews and repeated answer patterns are held for review. Nothing enters the result until a supervisor clears it.
- Cluster-level GPS checks
- Duration and pattern flags
- Back-checks on a fixed share
- Results only from cleared records
Choosing a study
Use the decision you face to pick the study.
| Study | Decision it supports | When | What you receive |
|---|---|---|---|
| Winnability survey | Which name gets the ticket | 6 to 12 months before | Recognition, favourability, issue fit for each shortlisted name |
| MLA report card | Whether the sitting MLA is in trouble | Any time in the final year | Satisfaction by area and issue, compared with the party's standing |
| Opinion poll | Where the seat stands | Campaign period | Preference with margin, issue ranking, turnout intent |
| Pre-poll tracking | Whether the campaign is moving voters | Two or three rounds | Change charts with sample sizes |
| Exit poll | Early vote share read | Polling day | Estimate released after the ECI window ends |
Our view on anti-incumbency
Across a state
Parties shortlisting candidates for dozens of seats need comparable numbers, not dozens of different reports.
01Same core questionnaire everywhereOne core
Same core questionnaire everywhere
Recognition, favourability and issue fit are asked identically in every seat, so a score in one district can be compared with another.
- Fixed core questions
- Seat-specific issue block
- Names rotated
Survey 3/6 · AmenitiesOffline
How is drinking water in your area?
- Comes regularly
- Comes sometimes
- Rarely / never
- Don't know
✓ Section saved on phone · autosave
BackNext
02State view with seat drill-downOne dashboard
State view with seat drill-down
Leaders see all seats together, sorted by margin and sample size, and can open any seat to see areas and issues.
- Seat ranking
- Area drill-down
- Sample sizes visible
3,000
Target homes
2,184
Visits
1,962
Completed
1,871
Accepted
Daily accepted surveysLast 12 days
03Seats surveyed in wavesPhased field
Seats surveyed in waves
Field teams move region by region. Each wave is reported when it closes so decisions do not wait for the last district.
- Wave-wise release
- Regional language teams
- Same supervision rules
Ram Kumar · shift on
26.8467 N · 80.9462 E · ±12 m
14
Received
12
Accepted
1
Review
1
Outside
Polling day and campaign tools
Exit interviews at sampled polling stations
Interviewers approach every nth voter leaving a random sample of stations and record the vote privately on the phone. Results stay embargoed until the Section 126A window notified by the Election Commission ends, which in multi-phase state elections is after the last phase.
- Random station selection
- Private vote entry
- Refusal counts by hour
Vidhan Sabha survey agency interviewer recording a voter's answers privately
Vidhan Sabha election booth management app
Volunteers receive booth-area assignments, log canvassing visits and post polling-day status on an offline Android app. We do not upload voter rolls and the app does not track individual voters.
- Assignments per booth area
- Polling-day posts
- Alerts to the war room
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
A strongB fairC weakD very weakNo report today
Vidhan Sabha election voter survey app
Your own short doorstep surveys, run by volunteers, with GPS and time on every record. This data is labelled separately from our neutral poll.
- Skip logic and validation
- Offline sync
- Excel and CSV export
Survey 3/6 · AmenitiesOffline
How is drinking water in your area?
- Comes regularly
- Comes sometimes
- Rarely / never
- Don't know
✓ Section saved on phone · autosave
BackNext
Vidhan Sabha election campaign management company support
Door-to-door canvassing coverage maps and a daily field report for the candidate. Rallies, advertising and social media are outside our scope.
- Coverage by booth area
- Daily field report
- War room dashboard
3,000
Target homes
2,184
Visits
1,962
Completed
1,871
Accepted
Daily accepted surveysLast 12 days
Not the right fit if
- You want seat projections for a whole state from a handful of seats
- You want exit poll numbers released before the notified time
- You want voter-roll based profiling of named people
- You want the questionnaire worded to favour a party
A good fit if
- You need seat-level numbers with honest margins
- You are choosing candidates across many seats
- You want the MLA, party and leader effects separated
- You want raw data and a method note
Political survey solutions
All political research services.
Learn moreOpinion poll survey
Sampled pre-election polls.
Learn moreExit poll survey
Polling-day interviews and analysis.
Learn moreMedia poll partner
Field partner for state news desks.
Learn moreराजनीतिक सर्वे
Political surveys in Hindi.
Learn moreContact
Share the state and seats.
Learn moreFAQ
Straight answers to the questions people ask most about this topic.
How many interviews does an assembly seat need?
A seat-level estimate commonly uses 600 to 1,500 interviews. If you need results by block, ward or social group within the seat, each part needs enough interviews of its own. We calculate size from the margin of error you want and share it with the proposal.
Can you survey every seat in a state?
Yes, in waves by region using one core questionnaire and one dashboard. Timelines depend on how many languages and field teams are needed; we set that out in the plan rather than promising a date before seeing the list of seats.
What is an MLA report card survey?
It measures how residents rate the sitting MLA's work, visibility and responsiveness, area by area and issue by issue, kept separate from party and leadership preference. It helps a party judge whether a seat's problem is the candidate or the government.
Are your assembly surveys party-neutral?
The method is the same whoever commissions the work: random household selection, rotated names, neutral wording and published margins. Each client's data is kept in its own workspace and never shared with another client.
India Nationwide, Hindi-first
Send the state, the seats and the decision you need to take. We will reply with a sample design and a wave plan.