Data Required

What real reports need

TSiMP turns verified public sources and field inputs into aggregate intelligence. This page lists exactly what we need to move a constituency from sample/demo output to real, evidence-graded reports.

Source categories

The six source layers

Official Government

Govt portals, CMO, departments, collector, Assembly/Parliament, schemes, ECI/CEO, local bodies.

Party

Party official sites, manifestos, press releases and official social handles.

Leader

Official public pages of MLAs, MPs, CM, ministers, local & opposition leaders, aspirants.

News

Telugu/English digital papers, political & district portals, public TV/YouTube reporting.

Social Media

Public posts/comments on FB, Instagram, X, YouTube and public Telegram channels.

Field

Verified village/mandal/booth-cluster feedback, public meetings, grievance records.

What we need from you

Inputs that unlock real reports

Verified public pages

Official public pages of the sitting MLA/MP, CM, ministers, parties and opposition leaders (Facebook, X, Instagram, YouTube). Public/official accounts only.

Manifesto documents

Confirmed manifesto PDFs/sources (e.g. the Telangana Congress 2023 Abhaya Hastham manifesto — currently a placeholder until the source is verified). No promises are recorded until verified.

Government scheme sources

State/district scheme portals and notifications, eligible-vs-reached figures, and grievance/implementation records.

Local field inputs

Verified village / mandal / booth-cluster feedback, public-meeting observations and grievance records — aggregate only.

Public news & social signals

Telugu & English news portals, district/local media, and public posts on Facebook, X, Instagram and YouTube.

Per report

Data requirements by report

ReportRequired sourcesMin evidenceMin field inputs
MLA Impact ReportOfficial Government, Leader, Newsreported3
MP Impact ReportOfficial Government, Leader, Newsreported3
Constituency Political Movement ReportNews, Social Media, Fieldreported5
Village-Level Public Mood ReportField, Newsreported8
Local Leader Influence ReportField, Social Mediareported6
Manifesto and Scheme Impact ReportOfficial Government, Partyverified4
Opposition Activity and Aspirant Watch ReportNews, Social Mediareported4
Custom Election Readiness ReportOfficial Government, News, Fieldreported10
Weekly Constituency Pulse ReportNews, Social Mediareported2
90-Day Leadership Correction ReportOfficial Government, Field, Newsreported8

Each report also defines optional sources, scoring categories, output sections and risk flags (see lib/engine/report-requirements.ts).

Compliance

Why TSiMP avoids individual voter targeting

Every input above is used at the aggregate level only — constituency, mandal, village, booth cluster, and community / profession / age-group sentiment. TSiMP does not build individual voter profiles, individual persuasion, or person-level caste targeting, and never uses private WhatsApp groups or unauthorized scraping. This keeps the platform an ethical governance mirror, not a targeting tool.

Phase 3A

Ready for verified source onboarding

The TSiMP intelligence engine, database model and scoring are in place. The platform is now ready to onboard verified sources and generate real reports. Real reports require:

  • A verified party manifesto (with promises extracted)
  • Verified official leader pages (MLA / MP)
  • Verified government & election-commission sources
  • Verified news and public social sources
  • Verified aggregate field inputs

First pilot unit: Kodad Assembly Constituency (AC-90), Suryapet district — with Nalgonda Parliamentary Constituency. Everything stays aggregate; nothing is treated as verified until its source is confirmed.

Phase 3B

Real DIIE data activation checklist

To switch the Demographic Impact Intelligence Engine from sample to real data for the first pilot — Kodad AC-90 — these verified, aggregate inputs are needed:

Demographic inputs

Aggregate group weights (farmers, youth, women SHG, welfare beneficiaries, rural/urban…). Verified aggregates only — no invented percentages, no individual data.

News / event inputs

Dated public news, government decisions, scheme updates and protests — each mapped to issue category, severity, evidence and responsibility.

Scheme inputs

Scheme delivery status with eligible-vs-reached figures and local-leakage risk.

Source verification

Each source confirmed (official > media > social); allegations evidence-graded and archived, never asserted.

Safety note. TSiMP does not identify or target individual voters. It analyzes public issues, demographic segments, news impact, scheme delivery and leadership credibility at aggregate levels only (constituency, mandal, village, booth cluster, and community / profession / age segments).

Phase 4

Real Kodad report activation

The onboarding pipeline is ready. Verified Kodad AC-90 inputs flow through validation → PostgreSQL → DIIE + MPIE reports:

  1. 1.Provide verified CSVs (demographic profile, news events, public issues, schemes, sources, field inputs) — each with verification status + source URL.
  2. 2.Automated validation blocks any individual names, phone numbers or private identifiers before import.
  3. 3.Data is imported to PostgreSQL (aggregate-only) and reports are generated on real data.

First pilot: Kodad AC-90 (Suryapet), with Nalgonda PC. Still aggregate-only — no individual voter data is ever collected or analyzed.

Have the sources ready?

Share verified public pages, manifesto and scheme sources, and we will move your constituency from demo to real reports.