What happened in every polling station last time, and the time before that
Booth-wise election result analysis tells a candidate, party unit or consultant where votes actually came from. Bharat Survey takes the Form 20 result sheets published for assembly and Lok Sabha elections, turns them into clean polling-station tables, maps each booth to its village or ward, and compares two or more elections side by side. You get strong, weak and swing booths on a map, an Excel workbook you can filter, and a short PDF report that says plainly what the numbers can and cannot tell you.
Form 20 is the final result sheet prepared at counting under Rule 56C of the Conduct of Election Rules, 1961. It lists votes for every candidate at every polling station.
After counting, the Returning Officer finalises the Form 20 result sheet for the constituency. State Chief Electoral Officer websites publish these sheets, usually as PDFs, one file per assembly constituency, for recent assembly and Lok Sabha elections. For a Lok Sabha seat, the sheets are organised by the assembly segments inside it.
Each row is one polling station: its number, the valid votes for every candidate, rejected votes, NOTA and the total. Postal ballots appear separately at the bottom. That is the whole dataset. It has no demographics and no voter names, which makes it one of the cleanest public records a campaign can study.
Many sheets are scanned, some run across dozens of pages with candidate columns split between them, and polling stations are renumbered between elections when new booths are added or merged. A careless merge across two elections compares different places under the same number. Most of our effort goes into getting that matching right.
We gather the result sheets for every election in scope, record the file source and date, and check each constituency total against the published ECI result.
Candidate names are standardised and each column is tagged with the party or independent status recorded by ECI, so the same party lines up across elections.
Polling stations are matched by building, locality and number, with splits and mergers recorded. Booths that cannot be matched with confidence are flagged, not guessed.
Each polling station is linked to the village or ward it serves using the polling station list, so results can be rolled up to places people recognise.
Strongholds, weak booths, swing booths and turnout change are calculated, mapped and explained in a short report with its limits stated.
Labels only help if everyone uses them the same way. We agree the definitions with you and print them in the report.
Both are useful. The choice depends on how much the constituency has changed and what decision you are making.
Last result, booth by booth
Two or more elections matched
| Deliverable | What it contains | Format |
|---|---|---|
| Clean result workbook | Votes, vote share, margin, NOTA and rejected votes per polling station, per election | Excel |
| Booth match sheet | How each booth maps across elections, with splits, mergers and unmatched booths flagged | Excel |
| Village and ward roll-up | Results aggregated to village or ward where polling station lists allow | Excel |
| Booth category list | Stronghold, weak and swing booths using the agreed thresholds | Excel |
| Result maps | Booth or village layers coloured by leader, margin and swing | PDF print and interactive web map |
| Written report | Findings, definitions, limits and a list of booths worth a field visit |
The mix can be trimmed or extended on the scoping call; turnaround and price are on quotation.
Print and web maps with booth and village result layers.
Learn moreVoter list in Excel with booth-wise age–gender analysis.
Learn moreSwing, turnout and survey analysis at constituency scale.
Learn moreExplore assembly constituencies and their past results.
Learn moreFind out why a booth moved, with verified interviews.
Learn moreStraight answers to the questions people ask most about this topic.
Form 20 is the final result sheet prepared by the Returning Officer at counting, under Rule 56C of the Conduct of Election Rules, 1961. It shows the votes each candidate received at every polling station in the constituency, plus postal ballots, rejected votes and NOTA. State CEO websites publish these sheets after the election.
Yes, with care. Polling stations are added, split and merged between elections, so numbers alone are not reliable. We match booths by building and locality from the polling station lists, record splits and mergers in a match sheet, and flag any booth we cannot match with confidence instead of forcing a pair.
Where the State Election Commission publishes booth-level or ward-level results for panchayat or municipal elections, we can include them. Publication varies by state and by election, so we confirm what exists for your area on the scoping call before quoting.
We agree the definition with you before counting. A common version is a booth where the leading party changed between two elections, or where the margin moved by more than a set number of percentage points. The definition and threshold are printed in the report so the list can be checked.
No. Form 20 reports votes per polling station, each covering anything from a few hundred to over a thousand electors. It cannot show how any person, household or community voted, and our reports do not claim to. For reasons behind a shift, a field survey with consent is the right tool.
It depends on the number of constituencies and elections, the quality of the source PDFs and whether you want maps and a written report. We scope it on a call and send a quotation with a delivery date. There is no fixed price list.
Open reference pages built from ECI, Census 2011, LGD and MCA records, with the source credited on every page.
Tell us the constituency and which elections to compare. We will check that Form 20 sheets exist, confirm the deliverables and send a quotation.
