Methodology

How FinSach works.

FinSach turns India's public market disclosures into plain English. Here's exactly how — because a site that asks you to trust it should show you the machinery.

Everything on FinSach comes from public sources, and we name them:

  • Institutional flows (FII/DII) — from the exchanges.
  • Bulk and block deals — from the stock exchanges' own disclosure feeds.
  • Earnings and corporate events — from published company filings and market data.
  • Prices, indices, and commodities — from established market data providers.
  • Market events (F&O expiry, RBI policy) — from exchange calendars and the RBI schedule.

We don't have secret data. We have public data, decoded — the same disclosures anyone can access, translated into something you can actually read.

When we can't get a number, we show a dash

This is the most important thing about how we work. If a data source is down, or a figure isn't confirmed yet, we show "—" — never a stale number, never an estimate dressed up as fact. A blank is honest; a wrong number pretending to be right is not. When you see a dash on FinSach, it means "we don't have this right now," and that's the truth.

We describe what happened — we don't predict what's next

FinSach explains disclosures: what was filed, what a deal was, what an event is. We do not forecast prices, we do not tell you what will go up, and we do not tell you what to buy or sell. Where we surface something forward-looking (like an upcoming earnings date), we report the event — never a guess about the outcome.

How we read delivery breadth

For each BSE session we compute a turnover-weighted delivery ratio — across all traded stocks, delivered quantity as a share of traded quantity, weighted by the value each traded — and also recompute it excluding the five most-traded stocks. When the two are close, the reading holds without the biggest names; when they diverge, the headline leaned on a few. The "one-month norm" averages the headline over the trailing 20 sessions, counting only the ones where those two figures were close, so a block-heavy day doesn't set the baseline; below 12 such sessions we withhold the reading. (Excluding the top five is robust — the signal is nearly identical at three or ten; the thresholds are calibrated from two years and may be revised.)

What we will never do

We will never invent data to fill a gap. We will never show fake activity or manufactured social proof. We will never give you investment advice we're not licensed to give. And we will never tell you a number is real when it isn't.

That's the whole method: public sources, named; honest gaps, shown as gaps; plain-English decoding; and no advice, ever. Don't trust us because we say we're honest — trust the machinery, and check our work.