Fact-Checking Policy
What checking a fact means here
At WEBWIZARD 360, checking a fact means tracing a claim back to something we can point at, rather than to a mood, a trend or another outlet’s paraphrase. A claim is not confirmed because it is everywhere, because it is plausible, or because someone confident said it on camera. It is confirmed when it rests on a primary source we have looked at ourselves. This page describes how that works, so you can judge how much weight to put on what you read.
Fact-checking is not a separate department that inspects finished stories at the end. It is built into how a story is reported and edited from the start, because a claim caught before publication is a correction avoided. It runs across everything we cover, from a market figure to an election result to a clinical finding to a transfer story.
Primary sources come first
Whenever we can, we go to the original: the official release, the court or regulatory filing, the published dataset, the full transcript or recording, the study rather than the press summary of the study. We read the source before we read the coverage of it, because the quickest way to inherit a mistake is to summarise somebody else’s summary. When we quote a document, we link to it so you can read past our excerpt and check that we represented it fairly.
Where a primary source is not public, we describe what it is and how we know its contents, as precisely as protecting a source allows. If a claim can only be traced to a second-hand account, we say so, and we do not upgrade it to the status of confirmed fact by writing around the uncertainty.
How many sources, and how independent
How much confirmation a claim needs scales with how much rests on it and how much damage it would do if wrong. A routine, uncontested detail from an official source can stand on that source. A serious allegation, a surprising number, or a claim someone has an interest in pushing needs more: independent corroboration from sources that are not simply repeating each other.
Two outlets citing the same anonymous briefing are not two sources, they are one source heard twice, and we treat them that way. We watch for the circular reference, where a claim bounces between reports until it looks established without anyone ever having verified it. When the corroboration is not there yet, the honest move is to wait, and usually we wait.
Numbers, quotes and images
Numbers get particular scrutiny because they carry false authority. We check a figure against its origin, we check the units and the timeframe, and we check that a percentage and its base actually describe what the sentence claims. A statistic detached from its denominator can be technically accurate and still deeply misleading, and catching that is part of the job.
Quotes are checked against the recording or transcript rather than reconstructed from memory, and we do not trim a quote in a way that changes its meaning. Images and video are checked before use: we confirm what an image actually shows, when and where it was taken, and that it has not been staged, altered or lifted from an unrelated event. An old photograph relabelled as new is a classic way misinformation travels, and we would rather run no picture than the wrong one.
Developing stories and provisional facts
Fast-moving news is the hardest test of any checking process, because the pressure to publish is highest exactly when the facts are least settled. Our rule is to be clear about the state of our knowledge. We label developing coverage as developing. We separate what is confirmed from what is reported but unconfirmed and from what is still unknown, and we attribute early claims to whoever is making them rather than adopting them as our own.
As better information arrives, we update the same story in place and note what changed, instead of leaving an early, weaker version standing. Early casualty figures, first-day estimates and initial official statements are treated as provisional by default, because they very often are, and history is full of first numbers that later moved a long way.
When we are not sure
Sometimes the honest answer is that we do not yet know, and we will say that rather than manufacture a certainty we have not earned. Not publishing a claim we cannot stand up is a decision we make often and do not apologise for. A claim held back for a day and confirmed is worth far more to you than a claim rushed out and retracted.
We also try to be transparent about the limits of a source. A single study is a single study, with the caveats that come with it. A poll is a snapshot with a margin of error. An official first account can be revised. Flagging those limits is not hedging. It is telling you how firm the ground actually is.
When a check fails after publication
No process catches everything, and when a fact we published turns out to be wrong, we fix it openly rather than hide it. The correction goes on the story with a dated note that says what was wrong and what the correct fact is, under our corrections policy. Where a checking process let something through, we look at why, because a repeated failure is a process to fix, not a person to blame.
If you believe we have published something that is not true, help us check it. Write to support@webwizard360.com with the link, the specific claim, and the source you think we missed. That is not a nuisance to us. It is part of how the work stays accurate, and it fits alongside the wider disciplines in our editorial standards.
Last updated: August 2026
