How it works
One page, one question: do the numbers on this watch fit what the public record says about the brand? The answer is a list of lines, each with a status, a confidence and its sources.
What the check does
You pick a brand, type the serial and, if you have it, the reference. The check runs on the server, so every lookup is counted, and answers with up to four lines.
- Format. The serial is matched against the patterns the brand has used, era by era: the number of characters, letters against digits, a prefix. The line says which era the format fits, or that it is unusual for the brand, or that it fits no pattern we know. Before giving up, the check tries the usual misreadings: O for 0, I for 1, S for 5, B for 8.
- Year. Where published tables assert years for that era, the line shows the range and names the tables. When tables disagree by more than the row's tolerance, the line shows both values rather than picking one. When the era has no rule, which is the case for every Rolex since spring 2010, most Omegas after about 2008 and all modern Tudor, the line says "not derivable" and points to the warranty card.
- Cross-check. With a reference, the line says whether the serial's era and the reference's production window fit each other. A reference that is not in our table yet says so.
- Reported issues. What owners have reported about the calibre or family, attributed to the forum and dated, with how to check. Never a fault stated as fact.
Under the lines sit the two lost-and-stolen registries. No lost-or-stolen check is performed on this site; the registries are linked so you can run one yourself.
The statuses
| Line | Status | What it means |
|---|---|---|
| Format | matches | The serial's shape fits a pattern the brand used in a known era. |
| Format | unusual | The length fits but the characters do not. Often a misread engraving; worth checking again. |
| Format | no match | The serial fits no pattern we know for the brand. |
| Year | range | Published tables agree within the row's tolerance; the range and the tables are shown. |
| Year | disagreement | Published tables differ; both values are shown with their publisher. |
| Year | not derivable | The serial carries no date for this era. The warranty card is the only reliable source. |
| Year | no table | No reviewed table for the brand yet. |
| Cross-check | fits | The serial's era and the reference's production window overlap. |
| Cross-check | does not fit | They do not overlap. Worth asking the seller about. |
| Cross-check | not in our table | The reference has no reviewed row yet. |
What a person reviewed
The tables behind the check are compiled from published sources and reviewed by a named person before they go live. A row carries its sources with the date they were read, a confidence of high, medium or low, the reviewer's name and the date of review. A row without a source and a reviewer never goes live; a draft the reviewer has not opened is never read by the site.
A model drafts rows from the sources a reviewer queues; it never writes to the live tables, and a row it cannot source is dropped, not guessed. The reviewer accepts, edits or rejects each row by hand. Corrections from readers go through the same path: a correction is a draft, not an edit.
Right now: 0 of 5 brands live, 0 reviewed rows, 0 sources. The publishers are listed on the sources page.
What the check never says
Whether a watch is real or fake. The check compares numbers with published tables; it has not seen the watch, and the tables are not the brand's own records. The honest line for a modern Rolex is "not derivable", and the page says it rather than guessing. Where a year matters, the brand's own archive extract, where the brand offers one, is the document to ask for.
Whether a watch is lost or stolen. That check belongs to Enquirus and The Watch Register, and the page links to both.