Established
Two or more newsrooms report the same fact from their own sourcing, and you can reach the report that published it first. Share the original report, not the post that led you to it.
Reader guide
How to verify a news story before you share it.
Find the original report, not the post about it. Then look for at least two newsrooms reporting the same fact from their own sourcing, and read what those reports establish rather than what the headline implies. If you cannot find the original within five minutes, treat the story as unconfirmed and do not share it.
The three outcomes.
Every check ends in one of three states, and the state tells you what to do next.
Disputed
Credible newsrooms agree on the broad story and disagree about a fact that matters. Share both accounts, or share neither. Do not keep the part you liked and drop the part you did not.
Unconfirmed
You cannot reach an original report, or every account you found traces back to one source. Wait. A check that ends in “unconfirmed” is a finished answer, not a failed one.
These are the same three states our own coverage preserves when accounts differ, and they are useful because they tell you what to do rather than how to feel about the story.
Step 1: find the original.
Trace the post back to the report that published the fact first.
A post is usually a summary of a summary. The original carries a byline, a publication date, and either named sources or a document you can inspect. When a post links to a newsroom, open that newsroom's page and read the story there. When a post links nowhere, search the exact wording of the claim and look for the newsroom that published it first.
Reading a page closely is not the same as checking it. Wineburg and McGrew, working with the Stanford History Education Group, recorded in 2017 how professional fact-checkers judged unfamiliar sources: instead of staying on the page in front of them, they opened other sources and read what those said about it. The researchers called the habit lateral reading, and on a phone it takes seconds.
Step 2: count independent confirmations.
Two reports that cite each other are one source, not two.
Independence means each newsroom did its own work: its own interviews, its own documents, its own named sources. A wire story republished on ten websites is one report. A second newsroom that credits the first is a relay, not a confirmation. When you count, count the newsrooms that stood the story up themselves, and treat the rest as copies of the same source.
AFP Fact Check’s guide “How we work” says it verifies a fact-check’s central claim with at least two independent sources.
Mike Caulfield set out the same discipline in 2019 as the SIFT method: stop, investigate the source, find better coverage, and trace the claim to its original context. The last two moves are the counting step, and they are the ones most readers skip.
Step 3: read what the reports establish.
Read the body of the report for the fact the reporter stands behind, not the wording of the headline.
Headlines are short, and short lines drop the qualifications that matter most. The body of a report usually says what the newsroom knows and how it knows it, and the difference between the two is where most misunderstandings start. Two newsrooms can agree on the broad shape of a story and disagree about the single sentence you were about to repeat.
Three phrases carry less than they appear to. “Officials said” names a person in authority as the speaker, not a fact the newsroom confirmed. “Reportedly” means the newsroom has not confirmed the claim itself. “Alleged” means no court has decided the question. None of the three is a lie, and none of them is a confirmation either.
Step 4: check the date and the place.
A real photograph or a real report from another year, or another country, can still be false when it is attached to this story.
Old images return with every new event: a flood, a protest, a wreck, an explosion. Check the date on the report, and check the dateline that gives the place it was filed from. Where the story is a photograph, look for the same image in an earlier report, and read the details a caption cannot fake: shopfronts, licence plates, street signs, and the weather.
If the date or the place does not match the claim, the story is unconfirmed no matter how many people have already shared it. Recycled material is an easy way to pass on something false without meaning to.
Step 5: decide.
Established: share the original. Disputed: share both accounts, or neither. Unconfirmed: do not share it.
Sharing the original report is better than sharing the post that led you to it, because the report carries the byline, the date, and the sourcing that a reader needs to check your work. When two credible newsrooms disagree, a reader who shares one side is repeating a claim that is still moving. When you cannot trace a claim in five minutes, waiting costs you nothing. Sharing it costs you the trust of the people who read you.
This check is deliberately small. It does not settle what happened, and it was never meant to; it tells you whether the story in front of you is ready to pass on.
The five-minute checklist.
Run these seven checks in order and stop as soon as one of them fails.
- Find the original report, and note its byline, its date, and the sources it names.
- Open the newsroom's own page, and judge the story there rather than in the post that shared it.
- Count only the newsrooms that did their own reporting, and set the relays aside.
- Read the body for the fact the reporter stands behind, and quote that fact.
- Write down what the reports disagree about, and keep the disagreement if you share the story.
- Check the date on the report and the place it was filed from.
- Decide, then share the original report, or share nothing at all.
A fixed example: April 2013 in Boston.
A crowd can build a confident story out of fragments, and the fragments may never have been checked against a source.
After the April 2013 Boston Marathon bombing, Reddit and 4chan users wrongly identified several people in marathon photos as suspects. The Guardian’s Charles Arthur reported on April 19 that the mistakes left innocent people fearful. CBS News’ Chenda Ngak reported on April 22 that Reddit apologized for fueling online witch hunts and had apologized privately to the family of missing Brown University student Sunil Tripathi, whose name and photo had circulated online.
This case shows why online repetition is not independent confirmation: look for newsrooms that reach the fact through their own reporting.
Sources: The Guardian, Charles Arthur, “Reddit’s Boston marathon speculation shows the limits of crowdsourcing” (2013); CBS News, Chenda Ngak, “Reddit apologizes for Boston ‘online witch hunt’” (2013).
Where Signal Brief fits
If your question is about the outlet rather than the story, the newsrooms behind our own coverage are named on our Sources page. For the method behind our editions, see how Signal Brief works.
This is the check we run before anything reaches a Signal Brief reader. Our coverage carries a receipt: the number of reports we checked and the newsrooms we used, so you can see the work behind it.