What a connection on ECHO's event graph actually means
A line between an event and a company is an invitation to investigate. It is not a complete explanation of what happened to that company's share price. ECHO's Live Map makes a news feed easier to explore by grouping topics and connecting headline mentions to tracked assets. Knowing how those links are built helps you ask better questions when you select a node.
Read the type of relationship before following it
ECHO's graph has a center, topic groups, event nodes and asset nodes. A topic connects to headlines assigned to that category. An event connects to a tracked asset when the headline matches its name or recognized alias. These are navigation relationships built from the available feed and quotes, not estimates of how strongly an event changes a market.
A short line, a nearby node or a cluster of headlines is not a measurement of economic exposure. Likewise, a quote displayed beside an asset is separate from the reason its headline connection exists. The matching process does not establish that the article explains the displayed price change. Read the relationship as 'this headline mentions this instrument' before adding any stronger claim.
A headline can support one connection and leave another open
Consider the illustrative headline 'Microsoft signs a solar power agreement.' It is invented for this guide and does not describe a reported transaction. If it appeared among the displayed events and Microsoft were among the available tracked quotes, the headline's company name could connect the event to MSFT. That connection would reflect the mention.
Suppose you then proposed a second link to a solar equipment supplier. The headline alone would not establish who supplies the project, whether equipment has already been purchased, or how large the contract is relative to that supplier's business. You would need the actual agreement, company disclosures or other evidence before describing an exposure.
Even a confirmed supplier relationship would not settle the share-price question. Revenue timing, costs, prior expectations and unrelated news could matter. Keep the supported relationship and the proposed market effect in separate sentences so the second cannot borrow the certainty of the first.
| Statement | Evidence needed |
|---|---|
| The headline mentions Microsoft | The headline text |
| A particular supplier receives an order | A document identifying the supplier and order |
| The order improves profit | Revenue timing, costs and contract terms |
| The announcement caused a price move | A separate causal analysis, not the graph line |
Turn a selected node into a source trail
Use the category controls and search to narrow the map. Select an event to inspect its headline and source, then open the event and underlying reporting where available. Follow an asset link when you want its market context. Make a note of the particular document that supports your interpretation rather than treating the whole cluster as one source.
W3C's provenance work distinguishes the information being produced, the activities that produced it and the people or organizations involved. That is a useful way to structure a research note: which document contains the claim, who issued it, and what transformation produced the summary you are reading? This is a research method, not a claim that ECHO implements the W3C PROV standard.
- Record the original source and publication time.
- Write the exact relationship supported by the source.
- Keep your proposed consequence visibly labeled as a hypothesis.
Do not treat an incomplete map as a complete market model
The map shows a selection of available events and matching tracked assets. Filters and display limits affect that selection. A company mentioned only in an article's body may have no link because matching uses the headline. An untracked supplier or an unavailable quote can also be absent. No visible connection is therefore different from evidence that no connection exists.
Names and aliases can be ambiguous, so check that a match refers to the intended company or asset. Several reports may also repeat one original announcement. More event nodes can mean more coverage rather than more independent evidence. Open the underlying sources before counting a cluster as confirmation.
Make the proposed mechanism explicit
Causal analysis asks what difference a change would make under specified assumptions. Judea Pearl's overview explains why causal questions require more than observed association. ECHO's topic and mention links are not a fitted causal model, and moving between connected nodes does not turn them into one.
A useful next question is: 'If this agreement is confirmed, which company disclosures would show who benefits, by how much and when?' Ask ECHO to help outline possibilities, then inspect the source documents yourself. Keep the known event, the candidate mechanism and the unresolved evidence distinct.
You can finish a session with a narrower question instead of a prediction. That is progress: you know which link is supported, which link you proposed and what would be needed to justify it. The graph has helped you navigate the research without becoming a substitute for the research.
Questions about a source or explanation? Read our editorial approach and report a correction. AI scenarios remain possibilities to investigate.
Explore a connection on the Live Map