Four Questions for Any Piece of Content

Provenance: is this the original item, or the copy that reached you? Find the earliest instance, not the loudest. Reverse image search across more than one engine, because their indexes and their strengths differ.

A miss is the part people get wrong: no result means the index holds no earlier copy, not that none exists. The original may sit on a closed platform, a messaging app, or a page nothing has crawled.

Source: who published it, what is their history, and what access would they need to have captured this? An account created last week is not a witness. Neither is an account with a long history posting something it could not possibly have seen.

Date: the post time is when it was posted, which is a different fact from when it was captured. Do not expect metadata to help — major platforms strip it on upload, so its absence on a downloaded image is the default condition and carries no signal. Corroborate instead with what is in frame: shadows, weather, foliage, signage, visible events.

Location: identify fixed features and match several of them independently. One match is a coincidence waiting to be discovered.

Then grade what you have, source and information separately, and find one corroborant that could have come out differently. Forty accounts carrying the same screenshot is one source amplified forty times, which is a fact about distribution and not about the world.