A screenshot arrives in the family group chat, forwarded twice already, with a line of capital letters at the top and a photo that looks convincing. It takes about thirty seconds to feel something about it and roughly the same amount of time to pass it on. Checking it properly takes longer — maybe ten minutes — but the routine is not complicated, and once you have done it a few times it becomes almost automatic.

What follows is the version of that routine used most often in newsrooms. It leans on three habits: finding the oldest version of a claim, searching the image rather than the words, and reading the document everyone is arguing about instead of the summary of it.

Step One: Freeze the Claim Exactly as You Found It

Before you check anything, capture it. Take a screenshot that includes the account name, the timestamp and the platform interface. Copy the URL. If the post contains a quote or a number, paste the text into a plain document rather than retyping it.

This sounds like busywork and is not. Posts get edited, deleted, or quietly changed after they travel. The wording drifts too. A claim that begins as "officials are considering" becomes "officials have approved" after two shares, and the second version is the one people remember. Having the original pinned down means you can say precisely what was claimed, and when.

Write one sentence describing the core assertion. Something like: a photo of a flooded street was taken this week in this town. Everything after that is a test of that sentence.

Step Two: Work Backwards to the Oldest Version

Most viral claims are not new. They are old ones with a fresh date attached, or a single account's interpretation of a genuine event. To find out which, trace the claim upstream.

Put the most distinctive phrase in quotation marks and search it. A string of unusual words will usually surface the original post, a press release, or the same claim recycled from an earlier year. Then look at the accounts sharing it. Is the poster the source, or are they sharing a screenshot of somebody else's post? Accounts that mostly repost other people's material rarely have information you cannot find closer to the origin.

Three quick tests for recency

  • Check the earliest timestamp you can find, not the one on the post in front of you.
  • Search the claim with an earlier year added. Recycled claims often turn up in a previous cycle, word for word.
  • Ask whether the account existed, or was active on this subject, before the event it describes.

If the oldest version you can reach is a screenshot of a screenshot, treat that as an unfinished trail rather than proof of anything.

Step Three: Reverse-Search Every Image and Video

Images are where most of the confusion lives. Run a reverse image search and crop tightly around the distinctive detail first. Text overlays and borders confuse the search, so cut them out.

Then read the image itself. Shop signs, number plates, road markings, plant life, uniforms and the language on a poster are all clues. Weather is another: a photo described as taken this week is easy to test against a forecast for that place and date.

For video, pull two or three still frames and search those, since a whole clip rarely matches. Check the watermark in the corner, the account that uploaded it first, and whether anyone in the footage is identifiable and speaking in their own words. First-hand footage usually comes with a messy background — wind, shaky framing, people talking over each other. Polished footage with a logo and a neat caption is usually a repost.

Step Four: Go to the Primary Document

Arguments about numbers, quotes and rules are settled by documents, not by the posts about them. If a claim says a council approved something, the minutes are public. If it quotes a report, the report is usually available in full. If it concerns a court case, there is a filing or a judgment.

Two things go wrong at this stage, and both are common.

The first is that the document does not exist. A claim attributes a figure to a study nobody can find, or a statement to a meeting that was never held. Absence of a document is weak proof on its own, but it is a strong reason to withhold belief.

The second is subtler: the document exists, and it says something close to the claim but not the same thing. A real report can be quoted out of context, a preliminary finding can be reported as final, or a proposal can be described as a decision. Read the section around the number. Look for the words "estimated", "draft", "subject to" and "in some cases". They change the meaning considerably.

If you cannot find the document and the claim matters, ask the person making it for the source. A specific request for a link is one of the most effective checks there is.

On medical, legal and financial claims, remember that a document being real does not make a self-treatment plan or a legal shortcut safe. Those questions belong with a qualified professional who knows the specifics.

Step Five: Compare Independent Reporting and Note What Is Missing

Look for two outlets that did their own reporting rather than repeating a wire story or quoting each other. Newsrooms make mistakes, but independent verification is still the quickest way to see whether a claim holds.

Pay attention to the gaps. If the only sources for a claim are screenshots, and no reporter, official body or witness has confirmed it, that is information in itself. Check the denials too. A correction issued quietly six hours after the original post is still a correction, and it rarely travels as far as the claim did.

When You Cannot Verify It, Say So Plainly

Not every claim can be settled before you have to decide what to do with it. Use three labels and be honest about which one applies: verified, unverified, or false. Add a fourth if needed — true but misleading — for accurate details arranged to create a false impression.

When you correct someone, lead with the finding and keep it short. "That photo is from a different city, here is the original" works better than a lecture. Share as little of the false claim as you can while still being clear, because repetition has a way of sticking.

A Five-Minute Routine You Can Keep

Here is the whole thing compressed. Keep it nearby until it becomes habit.

  1. Save the post, the URL and the exact wording.
  2. Search the distinctive phrase and find the oldest version.
  3. Reverse-search the image or still frames, cropped tight.
  4. Find the primary document and read around the quoted line.
  5. Check two independent reports, then label the claim verified, unverified or false.

You do not need to run this on everything. Choose the claims that would change what someone does — a health decision, a vote, a donation, a journey — and let the rest go. The point is not to become suspicious of everything. It is to be the person in the group chat who takes ten minutes, comes back with an answer, and is right often enough that people start waiting for it.

Photo: kaosnoff / Pixabay