Track stock mentions across a measured 48-hour window

What Mentioned measures

The current mapping follows 47 curated equities and 10 crypto assets. For each target, an exact configured entity-name or GKG-theme match in one valid GDELT source record contributes one mention. The same record can match multiple targets, while multiple matches for one target in that record still count once.

General company and entertainment coverage can count alongside financial reporting. A record about a Netflix show can count for Netflix if it matches the configured company name. Syndicated or substantially similar coverage can appear as separate records; the total is not a count of unique stories.

This is intentionally narrower than “everything said about stocks online.” It does not cover every listed security, asset, source, article, language, or conversation, and it does not identify unique people or unique events.

How to read the chart

A current view requires all 192 expected 15-minute source intervals in the latest 48-hour window. Mentioned groups every six intervals into a 90-minute bar, so each chart contains 32 measured buckets ordered from oldest to newest. The headline count is the sum of those bars. The window rolls forward, so older records leave the total as new intervals arrive. Each bar represents a 90-minute period, rather than the exact time of each reference.

Publication and collection delays mean this is not real-time. Use the displayed snapshot time to judge freshness. If any expected interval is missing or fails integrity checks, Mentioned withholds a new “current” result instead of converting the gap into a zero.

Why mention activity can be useful context

A consistent aggregate window can make it easier to see which names appear more often within the tracked universe and when those references were concentrated. That can help organize a research queue without exposing individual source items.

The count cannot tell you whether coverage is favorable, critical, promotional, duplicated, or important. It should be paired with independent research before any financial decision.

Source disclosure and privacy boundary

The mention source is the GDELT Project. Mentioned stores and displays aggregate per-entity counts and interval timing. This lane does not retain or display article text, article URLs, article titles, or publisher identities.

Premium can show up to four recurring, service-allowlisted aggregate topic labels that appear in at least two matched records across the complete window; Free responses strip them. The feed does not expose raw keywords or claim sentiment, causation, or a narrative behind a count. See the full methodology for exact counting and completeness rules.

How stock mention alerts work

Customer notifications are currently paused during prelaunch. Saving a watchlist or notification preference does not start delivery.

For supported tracked stocks, Mentioned compares the latest measured mention activity with that ticker's own recent baseline. Alert evaluation and delivery proceed only when the stock source window, alert authority, saved threshold, and delivery destination are current and valid; otherwise they fail closed. An alert reports mention activity, not price movement, sentiment, or investment advice. Crypto mention alerts are not available in this release.

A practical workflow

1. Save the tickers you already want to research.
2. Check the snapshot time and confirm the window is current.
3. Compare the total with the 32 timing buckets to see whether the references were concentrated or spread across the window.
4. Research the underlying coverage independently; Mentioned does not retain the source articles in this aggregate lane.
5. Make any financial decision from independent evidence, never a mention count alone.