Editorial standards

Who publishes this

SurnameGuessr is a small independent site. One person builds and runs it, with no newsroom behind them and no institutional affiliation. Articles are published under the site name rather than a personal byline, and the site takes responsibility for everything on it.

That means we cannot ask you to trust a set of credentials. What we can do is show our working, so you can check any claim yourself without taking our word for it. Every article carries a source list.

Where the surname data comes from

The origin country attached to each surname in the game starts with NamSor, a name-analysis service that classifies names by origin using models trained on large international name datasets. Names that come back with a weak or split signal get filtered out before they reach the game.

Frequency figures come from national statistics agencies wherever one publishes them. Spain’s INE, the UK’s ONS, Korea’s KOSTAT and Japan’s e-Stat all release surname counts, and those are better than any aggregator. Where no official count exists we use Forebears, which models global surname incidence, and we say so in the source list rather than presenting a modelled figure as a census count.

How articles get researched

Etymology is contested territory. A surname that looks obviously Spanish can turn out to be Basque, and plenty of popular origin stories fall apart under a small amount of scrutiny. So the rule for anything we publish is that a claim needs a source you can go and read.

• Every number in an article links to where it came from. If we cannot find a source for a figure, the figure comes out.
• Etymologies lean on reference works such as the Oxford Dictionary of Family Names in Britain and Ireland, and on academic and national-archive material for naming law and decrees.
• Where historians disagree, we say they disagree instead of picking the version that reads better.
• We do not invent example families, composite case studies, or illustrative anecdotes. Everything presented as having happened did happen, to real people, with a citation.

How we use AI

We use AI writing tools to help draft and structure articles. This is worth stating plainly, because a lot of sites use them quietly and it changes how you should read what follows.

A drafting tool is useful for turning research notes into readable prose. It is unreliable about facts, and it will produce a confident sentence containing a number it made up. So no AI-drafted claim goes live unchecked. A human verifies each factual statement against the source, rewrites the prose, and cuts anything that cannot be supported.

An earlier version of this blog was published without that check in place. We have since cut it from 86 articles to 35, removed duplicated topics, and gone back through what remains to verify the numbers and attach sources. Every article now carries an “Updated” date recording when it last went through that check, along with the sources it rests on.

Corrections

Tell us when something is wrong and we will fix it. Email surnameguessr@gmail.com with the article or the surname, and a source if you have one.

Corrections to an article update its “Updated” date. Disputed surname origins in the game get reviewed and either corrected or pulled from the pool. Family knowledge counts here. If your own name is filed under the wrong country, you probably know better than a classification model does.

What we are not

SurnameGuessr is not a genealogy service and cannot trace your family tree. Knowing that a surname originated in one country tells you close to nothing about where any particular family carrying it came from, and we try never to blur that line. For actual ancestry research you want records, not etymology.