Methodology

How This Tracker Works

Data sources, collection methods, confidence scoring, stage definitions, and limitations — so you can judge the reliability of every record.

Core Principle

This tracker uses only public data. No confidential employer data, no proprietary research, no insider information. Every record is sourced from a publicly accessible document or portal, and that source is recorded alongside the data.

This tracker is not affiliated with any government agency, OEM, STU, bus operator, charging company, or financial institution. It is an independent public-source intelligence service.

1. Data Sources

Every record names its source. Sources fall into two groups: those with live ingestion contributing data today, and those that are mapped but not yet active. We list both honestly so you can judge coverage rather than assume it.

This table is generated from the source-coverage registry, not written by hand — a source's group is derived from its ingestion mode, so a planned source cannot appear as active.

Currently active sources (4)

These are the only sources currently contributing data to the tracker:

Source Ingestion Confidence Last crawl Known gaps / caveats
BSE Corporate Announcements Automated High 2026-08-08 03:51:13 Retention window ~2 years; older filings need manual backfill.
CESL Tender Portal Automated High 2026-08-08 03:51:18 Corrigenda can overwrite originals; documents are archived on capture to mitigate.
Trade press (ET Auto, Mercom, etc.) Manual review Medium not yet recorded Good for large awards; smaller state tenders and corrigenda coverage incomplete.
User-reported source Manual review Low not yet recorded Submitted via the report form; surfaced as user-reported in the source trail when resolved.

Planned / not yet active sources (10)

These sources are mapped but not yet contributing any data. They are listed for transparency about intended coverage — no record currently comes from them, and the tracker does not cover what they would provide until they are activated.

Source Status Confidence (when live) Notes / known gaps
APSRTC (Andhra Pradesh) tenders Planned Medium Not yet indexed.
BEST (Mumbai) tenders Planned Medium Several BEST tenders run via Mahatenders / CESL lots. Not yet indexed.
BMTC (Bengaluru) tenders Planned Medium Karnataka e-proc carries bid submission. Not yet indexed.
CPPP (eprocure.gov.in) Planned Medium Broad discovery, weak on corrigenda/awards which live on issuer portals. Not yet indexed.
Delhi Transport Corporation tenders Planned Medium DTC publishes tender notices publicly on dtc.delhi.gov.in/tenders (notice board, no login) and on the Delhi NIC eProcurement portal govtprocurement.delhi.gov.in (public Latest Tenders/Corrigendum listing, ~10-item rolling windows). Bid submission is DSC-gated but discovery is open and no captcha is evident, so it does NOT defer like GeM/state-eproc. Recon done 2026-06-26; no scraper built and no rows ingested yet; comprehensive historical discovery not yet validated. Corrigenda are NOT captured: the Latest-Corrigendums feed identifies tenders by issuer NIT ref, a different namespace from the Latest-Tenders feed's GePNIC system codes, so corrigenda cannot be linked to their parent tender by ref; deferred until a linkage mechanism (e.g. PDF-level parent-id extraction) exists.
GeM (Government e-Marketplace) Deferred Low DEFERRED. Login-gated, bot-defended. Explicitly a known coverage gap for now.
Lok Sabha / Rajya Sabha Q&A Planned Medium High-credibility sanction/deployment figures. Keyword-triggered; pre-2022 not indexed.
State e-procurement portals Deferred Low DEFERRED. DSC-gated bid systems (Mahatenders, Delhi/Karnataka/AP/TS e-proc). Known gap.
TSRTC (Telangana) tenders Planned Medium Not yet indexed.
Vahan Analytics Dashboard Planned High Scraper in progress. State data revises ~3 months; trailing months re-captured.

2. Collection Method

Automated collection

Automated scripts collect data on a daily cadence from 2 sources (derived from the registry in §1):

  • BSE Corporate Announcements
  • CESL Tender Portal

Any source shown as planned in §1 (e.g. Vahan registration analytics) is not yet active — no data is collected or shown from it today.

Manual data entry

The following are added by manual review with cross-verification: Trade press (ET Auto, Mercom, etc.) and User-reported source. Alongside these, deployment history, OEM profiles, and state-level aggregates are manually curated from public annual reports, press releases, and tender documents. Each record carries a last_checked_at date.

Triage and staging

BSE announcements are automatically staged for review. A triage pipeline classifies each announcement as: order (bus count confirmed), press release (deployment update), tender (new or corrigendum), or boilerplate (financial filings, routine disclosures). Only classified non-boilerplate items are surfaced.

3. Procurement Stage Definitions

These terms have precise meanings in the context of this tracker. Do not treat them interchangeably.

Stage Definition Data needed to confirm
Sanctioned Government has approved a programme budget and bus allocation for a state/city. Government scheme notification, press release from MHI/MoRTH
Tendered A formal RFQ/RFP/tender has been issued for bus procurement or operations. Tender document from official portal (CESL, GeM, state STU)
Awarded A bus operator or OEM has been selected via bid evaluation. Award announced publicly. Official award notification or OEM BSE disclosure
LoA Issued Letter of Award issued to the winning bidder. BSE disclosure, press release, or tender portal update
Delivered Buses physically handed over to the operator at a depot. OEM press release, government handover event, or operator confirmation
Registered Buses registered with the state RTO and visible in Vahan database. Vahan data
Operational Buses in service on public routes, confirmed by operator or credible source. Operator press release, media coverage with specifics, government report
Cancelled Tender, order, or deployment officially cancelled. Official tender portal update or OEM disclosure

Records in this tracker only use "operational" when that stage is confirmed. Most deployment data reflects "delivered" or "registered" status. Unconfirmed stages are labelled with their confidence level.

4. Tender Status Logic

Tender statuses are computed as follows:

  • Open: Bid deadline is in the future and no award or cancellation is recorded.
  • Closing Soon: Bid deadline is within 7 days.
  • Extended: A corrigendum has extended the original bid deadline — confirmed by source.
  • Needs Review: Bid deadline has passed but no award, extension, or cancellation is recorded. This does not mean the tender is invalid — it may have been extended without public notice yet.
  • Awarded: Award data confirmed from official source or OEM disclosure.
  • Cancelled: Cancellation confirmed from official source.
  • Closed: Bid deadline passed and the process is confirmed complete (no award made public).

Expired tenders are never shown as Open. If a bid deadline has passed without confirmed extension, the status is set to Needs Review — not Open.

5. Confidence Scores

Every record carries a confidence label. This reflects how reliable the underlying source data is — not whether the event happened. A low-confidence record may be accurate; it just came from a source that was harder to verify.

High

Official source (government portal, stock exchange filing, tender PDF). Data is specific, recent, and directly verifiable.

Medium

Credible secondary source (trade press, OEM press release, annual report). Some interpretation required or cross-verification pending.

Low

Indirect source, incomplete data, or unverified inference. Use with caution — check original source before relying on this data.

Unknown

Source not yet reviewed or not publicly available. Field value may be estimated or derived.

Confidence scores (0–100) are also assigned where possible. These are heuristic — they do not represent statistical certainty. A score of 85 means the record is likely accurate based on source quality and recency.

6. Limitations

  • This tracker covers publicly available data only. Confidential terms, LOA amounts, or undisclosed extensions will not appear.
  • Only 2 sources are collected automatically: BSE Corporate Announcements and CESL Tender Portal. Trade press (ET Auto, Mercom, etc.) and User-reported source are added by manual review. Every other source listed in §1 is planned or deferred and contributes no data yet.
  • Registration data is not yet available — the Vahan source is planned, not active. When it goes live it will carry a known 1–2 month reporting delay and may undercount fleets that register in batches.
  • Deployment data relies on public disclosures. Not all cities with e-buses have made public announcements.
  • State STU tender portals (DTC, BEST, BMTC, APSRTC, TSRTC) are not yet scraped. Tenders from those bodies appear only when added manually as discovered, until their scrapers are activated.
  • OEM bus counts are "known deployed" from public sources — actual fleet sizes may differ from total produced or registered.
  • Corrigenda and deadline extensions are sometimes published only on tender portals without press coverage. Check source portals for latest status before making decisions.

7. Disclaimer

This tracker uses public sources only. Data may contain delays or errors. Always verify against original source documents before commercial decisions — including bidding on tenders, making investments, or entering into supply agreements.

Not affiliated with any government agency, OEM, STU, bus operator, charging company, or employer. Any similarities to proprietary research are coincidental.

The maintainer of this tracker is a private individual. Data is provided as-is, with no warranty of accuracy or completeness. The tracker is not responsible for any decisions made based on data published here.

Found a data error? Know of a tender not listed? Contact us to improve the data quality.

Subscribe to weekly digest