Alternative Data in Indian Markets

"Alternative data" means any data source beyond standard financial statements and price/volume data. In US markets, the alt data ecosystem is mature | satellite imagery of parking lots, credit card transactions, web scraping, job postings, and dozens of other signals are commercially available. In India, the ecosystem is earlier stage and more constrained by data availability and regulation.

This module maps the alt data landscape for Indian equities honestly | what exists, what's accessible at different cost levels, and how to evaluate whether any specific alt data source has edge worth pursuing.

The Indian alt data landscape

The underused free signal: promoter pledging

Of all alt data available for Indian markets, promoter share pledging is the most underused relative to its predictive power. Promoters pledging their shares to take loans signals financial stress | they need liquidity but can't or won't sell equity. High pledging (above 30 to 40% of promoter holding) consistently precedes corporate stress events in Indian markets.

• DHFL: promoters pledged heavily before the NBFC crisis hit. Pledging data was public months before default.
• Eveready Industries: promoter pledging rose sharply in 2018 to 2019 before the company ran into severe financial difficulties.
• Suzlon Energy: multiple rounds of high pledging correlated with periods of extreme financial stress and debt restructuring.

Pledging data is available for free on BSE/NSE. Filtering strategies to exclude stocks with promoter pledging above 25% improves quality and reduces tail risk substantially.

How to evaluate whether alt data has genuine edge

Any alt data source needs to pass four tests before you invest time building it into a strategy:

• Is there economic intuition? Why should this signal predict returns? If you can't construct a plausible mechanism, the backtest result is probably noise.
• Is the data truly available at signal time? Many alt data signals suffer from look-ahead bias | in hindsight, the data was available; in practice, there was a lag. Always check exact publication dates.
• Does the edge survive costs? Alt data signals are often short-horizon | they decay quickly. Short-horizon signals require frequent trading, which generates costs that eat the edge.
• Will the edge persist? Once a signal is known, it gets arbitraged away as more capital chases it. Alt data signals with low barriers to adoption decay faster than fundamental factor premia.

Three free Indian alt-data signals worth tracking

Most retail investors think alt data means satellite feeds and credit-card panels. The reality is that the highest-edge alt signals in India are free, public, and ignored because they require some assembly. Three worth pulling.

1. SHP filings (Shareholding Pattern, quarterly). Every listed company files an SHP within 21 days of quarter end. SEBI prescribes the format. The structured fields disclose promoter holding, FII holding, DII holding and individual large shareholder positions over 1 percent. The signal: rising FII or DII stake on a quality name, with promoter holding stable, is a positive concurrent indicator. Falling promoter holding paired with rising pledged percentage is the single strongest distress signal you can find in public Indian data. Both BSE and NSE publish SHP files in machine-readable format. Build a parser once and you have a quarterly screen running for free.

2. Bulk and block deal data (daily). NSE publishes bulk-deal data (any single trade above 0.5 percent of equity) and block-deal data (negotiated trades on the special block window) every trading day. The alt-data play: track which buyers are accumulating which names, and at which prices. A buyer who repeats across multiple bulk deals in adjacent quarters is signalling structural conviction. The data is free, the URL is public, the parsing is one CSV per day.

3. Insider trading filings (event-driven). SEBI requires designated persons (promoters, directors, KMPs) to disclose any trade above INR 10 lakh within 2 trading days. The disclosures are filed via NSE and BSE in the Insider Trading window. The signal: clusters of insider buying across multiple insiders within a short window have historically preceded outperformance over the following 6 to 12 months. Insider selling is noisier (could be tax planning or estate liquidity), but cluster buying is a strong positive signal in academic studies of Indian markets.

What does NOT work as alt data in India yet

Two kinds of alt data that work in the US and crush in India because the underlying ecosystem is different.

Credit-card transaction panels. The US has decade-deep card transaction databases that feed retail and consumer-facing equity research. India runs heavily on UPI, which is bank-to-bank, and on cash. Aggregated UPI data exists at NPCI level but is not granular at merchant-name level for outsiders. Card data alone misses most of the spending. Wait until UPI-merchant aggregation matures before betting capital here.

Satellite imagery on commercial real estate or factory output. These work in the US because companies tend to operate in single locations with verifiable footprints. In India, manufacturing is scattered, contract-based, and often shared with peers. The signal-to-noise of an Indian satellite feed for a typical mid-cap is poor. Satellite works best for ports, airports, large mining operations and a handful of integrated manufacturing facilities; it does not generalise across the broader market.

Glossary