A story does not have to be false to mislead. It can be true and misattributed, accurate and outdated, or technically correct and framed past recognition. Fact-checking is not about catching lies; it is about knowing what a claim actually rests on before you pass it on. The method below takes five minutes for a normal story and rewards every minute of it.
Step one: find the origin
Scroll past the aggregator to the earliest credible report of the claim. Who published first, and what source did they name? "According to a person familiar with the matter", "citing court documents", "the ministry said in a statement" — those attributions are the claim's foundation. A story that names no source and cites no document is passing along rumour regardless of how confidently it is written.
Step two: distinguish the kinds of claim
Not all sentences in a news story carry the same weight:
- Direct observation — a reporter at the scene, a photograph taken there.
- Documented fact — a filing, a record, a dataset anyone can consult.
- Official claim — what a government, company or party says about itself. True, false or somewhere between, it remains a claim.
- Anonymous sourcing — what officials said without being named. Weight depends on the outlet's track record and whether rivals confirm it independently.
- Analysis and speculation — what might follow. Often labelled; sometimes not.
Most viral misinformation is not fabricated — it is one category dressed as another: an official claim circulating as verified fact, or analysis presented as documented fact. Sorting the categories is most of the job.
Step three: check the image or video
Reverse image search is the fastest tool in the kit. Right-click an image and search it on Google Images or TinEye: if the same picture appeared three years ago attached to a different event, the match settles it. For video, look for the watermark of the original uploader, weather and shadows consistent with the claimed time and place, signage and vehicle plates consistent with the claimed location, and whether any established outlet is using the footage with its own verification note.
Step four: check the date and the place
Recycled news is quieter than fake news and spreads further. A genuine two-year-old story about a border dispute, re-shared during a new escalation, reads like breaking news. The checks: find the story's original publication date (aggregators strip them), search the headline to see whether outlets have updated or walked it back, and confirm the location's spelling matches the event — place-name confusion between similarly named towns is a recurring genre of error.
Step five: look for independent confirmation
One wire story republished by twenty outlets is one source. True confirmation means a second organisation verified the claim through its own reporting — its own officials, its own documents, its own imagery. When an event is major, the pattern to look for is multiple outlets with separate sourcing arriving at the same facts. When that pattern is missing, the honest summary is "reported, not yet confirmed".
Step six: read the correction history
Established outlets correct themselves visibly: a note at the top or bottom of the article, a corrections page, an editor's note. A visible correction history is evidence of a functioning process. Check the outlet's corrections policy — if it has none, treat its certainty accordingly.
The sharing test
Before sharing, one question: could I tell someone where this claim comes from? If the answer is "I saw a headline", the story is not ready to travel. If it is "the finance ministry announced it in a statement dated yesterday, and two independent reporters confirmed the numbers from the budget document", share it freely — with the date attached.
The primary-sources guide covers step one in depth, and our daily reports and analysis show the method applied to live stories, with every source listed so you can check the work.