Methodology & corrections

What we cover

World news, with resources, energy and critical minerals as the centre of gravity rather than a boundary. Oil and gas coming online, mines changing hands, export bans and processing mandates, the corridors and finance that move it all — but also grids, food, water, trade barriers and the rest of what decides who gets what.

A story has to pass two tests, not one:

• Is it lesser-reported? This is where we are actually useful. Bigger, faster newsrooms already own the saturated stories.
• Does it land on someone? Whose bill, whose supply, whose job, whose food. A story that is genuinely obscure but affects nobody does not run.

How we judge coverage

It is a measurement, not an editorial hunch. GDELT indexes worldwide article volume in near real time, so we can compare a story's significance against how widely it is already being carried. High significance, low coverage is what we look for.

Two things stop that number from lying to us:

• The wire test. A single agency report mirrored across hundreds of local papers looks thinly covered by outlet count but is in fact freely available everywhere. We measure outlets against distinct write-ups and drop anything that reads as syndicated wire copy.
• The placement test. Outlet counts are a signal, not a verdict — where a story sat matters more than how many sites ran it. A story filed in a section page is one almost nobody found; a story on a front page or in a push alert is one everybody did. We check what the major outlets featured, not merely what they published.

Widely covered stories are not banned. They run when we have a frame, a figure or a map the coverage missed — and they stay a minority of what we publish, because we cannot beat a wire service at reporting the news itself.

What we require before publishing

• Corroboration across independent sources. Nothing ships on one source alone without being labelled as such. Two domains are not automatically two sources — we check whether the texts are near-identical, which means one outlet plus a republisher, not independent confirmation.
• Single-outlet stories are marked on the card itself, in the post, not buried in a caption. If you see single outlet on a slide, one newsroom is the origin of that claim and it has not been independently confirmed.
• Allegations about named people carry their attribution in the post. If a claim about an individual cannot be stated as "court filings allege" or "prosecutors say" on the card itself, we do not run it. Where such a claim rests on a single outlet, we drop it outright.
• Neutral language. Loaded and emotive framing is stripped from source copy, and what was removed is recorded in our own audit.
• Every fact keeps its source URL in an audit file written at the same time as the post.

What we reject

Most candidates do not survive. The usual reasons:

• It was already everywhere and we had no frame, figure or map to add.
• The sources were the same article on two domains.
• The source text was too thin to establish what actually happened.
• It was routine: scheduled diplomacy, earnings, an opinion column, or a local-only item with no wider consequence.
• It was an untested allegation about a named individual.

Corrections

We will get things wrong. When we do, we correct the post publicly rather than deleting it, and the correction is logged here with the date, what was wrong, and how it was fixed. An account that never publishes a correction is not an account that never erred.

If you believe something we published is inaccurate, tell us — the fastest route is a direct message on any of the accounts below, with a source.

Who we are

Veritas Now is an independent, self-funded project. It takes no advertising, no sponsorship and no funding from any government, company or organisation with an interest in the sectors it covers. If that ever changes, it will be declared on this page before it affects a single post.