Method

Exactly what this page counts, so you can decide how much to trust it.

Sources

Three public RSS feeds are read per symbol: Yahoo Finance's per-ticker headline feed, Nasdaq's per-symbol outbound feed, and a Google News query for <SYMBOL> stock. Google News aggregates many publishers, so the outlet shown is the underlying publisher where the feed exposes it.

Relevance filter

Yahoo's per-ticker feed leaks general market stories that never mention the company, so a headline is only counted if it names the ticker ($SYM or the bare symbol) or the company name. Nasdaq's per-symbol feed is genuinely scoped and is trusted as-is. This trades a little recall for counts you can actually believe.

De-duplication

The same story reaches several feeds. Titles are normalised (lowercased, punctuation and whitespace stripped, first 70 characters) and duplicates are dropped, keeping the earliest copy. This under-counts rewrites that change the headline wording, so treat counts as a floor.

Velocity

velocity = headlines in the last 24h ÷ (headlines from 24h–7d ago ÷ 6). The denominator is the ticker's own recent daily average, so a mega-cap that always gets 30 headlines a day and a small cap that usually gets one are on the same scale. Where there is no usable history the ratio is shown as –.

Known limits

• RSS is delayed. Feeds refresh on their own schedule, and this page caches for five minutes. It is a research tool, not a real-time feed.
• Feeds cap how far back they publish, so the 7-day window is a sample, not a census.
• Headline count is attention, not direction. Ten headlines about a fraud investigation and ten about a buyout look identical here.
• No sentiment scoring, no price data, no recommendation. Nothing here is investment advice.