nameEditorial Policy
A name generator is only worth using if you can trust what is underneath it. This page explains how ours are researched and built, where we draw the line between real naming rules and invented ones, and what happens when we get something wrong.
Who is responsible for this site
Everything published here is researched, written and built by Sam Hollis, who runs the site. There is no content farm behind it and no rotating cast of anonymous freelancers. One person makes the calls, and that person can be emailed at [email protected].
How we choose which generators to build
Not every idea becomes a page. A niche has to clear three bars before we spend days on it.
- People are actually looking for it. We check search demand for the niche, and we read what people ask around it — on forums, in subreddits, in the mail we get. A generator nobody needs is a page nobody reads.
- We can do it meaningfully better. If the existing tools for a niche are a single button and a syllable pool, that is an opening. If a niche is already covered well and we have nothing to add beyond a different colour scheme, we skip it. Volume for its own sake is how a useful site turns into a content farm.
- There is enough naming material to work from. Some settings have a rich documented naming system. Some have four canon names and a vibe. We will still build the second kind, but we say on the page that the pattern is derived rather than documented.
Reader requests jump the queue. A large share of the catalogue exists because someone emailed and explained what they were trying to name and why nothing else worked.
How a generator gets built
Every generator follows the same five steps, in this order. The order is the point: research decides what the filters should be, not the other way round.
- Research the naming system. Before a single name is generated we work out how the setting or culture actually names people, places or ships. What are the parts of a name? Is there a patronymic, a clan marker, an honorific, an earned nickname? Are names meaningful words, or pure sound? A tradition with rules gets those rules written down first.
- Collect the raw material. Roots, prefixes, suffixes, epithets, titles and name elements, grouped by whatever the research said matters — gender, region, subrace, era, role. This is the part that takes longest and the part most tools skip.
- Design filters around the research. If Old Norse names carry a patronymic and an earned byname, the generator gets controls for both. If Khajiit honorific prefixes carry meaning, the generator shows the meaning. Filters exist to expose the naming system, not to pad the interface.
- Write the page. Each page explains the conventions the generator is using — so you can follow them, or break them on purpose, which is the more interesting option. Examples, common patterns and a FAQ come from the same research.
- Test the output. We generate at length and read the results. If a filter produces sludge, unpronounceable strings, or the same name twice a minute, the data gets rebalanced before the page ships.
Authentic rules vs. plausible fiction
This is the distinction we care most about, and it changes from page to page.
Real-world traditions — Old Norse, Japanese, Germanic, medieval European — have documented structures, and we follow them. A Viking name built here uses a genuine patronymic (-sson / -sdóttir) because that is how the system worked, and it never inherits a family surname, because those did not exist. We say so on the page.
Fictional settings split in two. Some — Elder Scrolls races, Tolkien's languages, established D&D lineages — have published naming material, and we work inside it. Others have a handful of canon names and nothing else. There, we derive a pattern from the names that do exist and generate in that style. That is informed invention, and we label it as such rather than calling it canon.
What we never do is generate a random syllable soup and describe it as authentic. If a page does not tell you which side of that line it is on, that is a bug — report it.
Sources
For real-world naming we use published grammars, etymological dictionaries and academic onomastics references. For fictional settings we use official material first — published rulebooks, in-game name lists, developer documentation — and community reference wikis only where nothing official exists, which we treat as weaker evidence. Where sources disagree, we go with the better-documented reading and, where it matters to the result, mention the disagreement on the page.
How the word lists and rules are built
Under every generator is a data file, and it is not a scraped list. Each one is assembled by hand from the research: root words and their meanings, prefixes and suffixes, epithets, titles, clan and place elements — each tagged with the things the naming system actually cares about, such as gender, region, subrace, era or social role.
The rules that combine them are written separately from the words. That separation matters: it is what lets a Viking name assemble as given name + patronymic + earned byname while a Khajiit name assembles as honorific prefix + given name, using the same engine but obeying two different systems. Where a combination would be wrong — a suffix that only attaches to one class of root, an honorific that does not apply to a given role — the rule blocks it rather than leaving it to chance.
Pools are then sized for variety. If a filter can only produce a few dozen distinct results, you will see repeats within a minute, so we widen the data until the output stops feeling like a loop.
Originality and protected names
Names are generated by recombining roots and patterns, not by sampling a list of existing characters. That is a deliberate design choice with a legal edge to it as well as a creative one.
- No canon characters in the pools. Data files hold roots and elements, not the names of published characters. A Star Wars generator is built from the phonetic patterns of a naming tradition, not from a list of characters in the films.
- Trademarked and protected terms are excluded. Faction names, product names and other trademarked terms do not go into the generation pools. They may be mentioned in the page text where it is genuinely explanatory — that is nominative reference, not output.
- Collisions get fixed. Recombination can occasionally land on an existing name by coincidence, especially in a small pool. Report one and we widen the data or add a block. This has happened, and it is a data bug, not an argument.
- We are not a rights clearance service. Names here come with no attribution requirement, but if you are naming a commercial product rather than a character, do your own trademark search. A generator cannot do that for you.
Fan-content niches are covered under the ordinary understanding that the underlying settings belong to their rights holders. This site is not affiliated with, endorsed by, or sponsored by any game publisher, studio or franchise owner named on it.
Human review, and where AI fits
AI assists here. It is not what makes the site work, and it does not decide anything that matters.
Where it is used: the hero and scene artwork on each page is generated with an AI image model from a written brief, then selected, cropped and colour-checked by hand — commissioning a hundred original illustrations is not something a free site can fund, and stock art would look worse. It is also used the way a competent writer uses any tool: drafting assistance, rephrasing, catching a clumsy sentence.
Where it is not: the naming research, the data files, the combination rules and the generator logic are built by hand. A language model does not know whether a byname pattern appears in the sagas — it knows what such a claim usually sounds like, which is exactly the failure mode this site exists to avoid. Nothing is published on the strength of a model's say-so about how a naming system works.
Every page is read by a person before it ships. That review covers four things: the naming claims are checked against the research; the generator is run at length and its output read for sludge, repetition and unpronounceable strings; the examples on the page are real generator output, not invented for the article; and the text is edited for accuracy and voice. A page that fails any of the four does not go live.
Corrections
We get things wrong. When we do, we fix the data or the text rather than quietly deleting the page, and substantive lore or linguistic corrections are made as soon as they are verified. Email [email protected] with the page URL and what is wrong. Corrections from people who actually speak the language or run the game are the most useful mail we get.
Review schedule
Generators are revisited when a source is corrected, when a setting publishes new material, or when reader feedback flags an issue — and the whole catalogue gets a pass at least twice a year. Each page carries the date it was last reviewed so you can see how fresh it is.
Advertising and independence
The site is free and funded by display advertising. Advertisers have no say in what we build, which niches we cover, or what a page says. We do not publish paid guest posts, paid links, or sponsored generators. Where a page links out to a game or publisher, it is because it is genuinely relevant, and those links are not paid placements.
Privacy
Generators run entirely in your browser. What you generate is not sent to us, stored, or logged. Details of the analytics and advertising cookies the site does use are in the privacy policy.
Editorial FAQ
Are the names on this site historically or canonically accurate?
It depends on the generator, and each page says which it is. Where a real tradition has documented rules — Old Norse patronymics, Japanese kanji meanings, Elder Scrolls honorifics — we follow those rules and explain them on the page. Where a setting has no published naming system, we build a pattern that fits its sound and label it as plausible fiction rather than canon.
Do you copy names from books, games or real people?
No. We work from roots, affixes and structural patterns, then recombine them. That is deliberate: it keeps results original and usable, and it avoids handing you a character name that already belongs to someone else's story. If a generator ever produces a well-known existing name, it is coincidence from a small root pool, and we will happily adjust the data if you report it.
Where does the naming data come from?
Published grammars and etymology references for real-world languages, official lore and in-game name lists for fictional settings, and community reference material where it is the only documentation that exists. We prefer primary sources over aggregator sites, and we prefer to state uncertainty rather than paper over it.
Do you use AI to make this site?
AI assists, but it is not the engine. The hero and scene artwork is generated to a written brief and then selected and edited by hand, and AI is used for ordinary drafting help on text. The parts that decide whether a generator is any good — the naming research, the word lists, the combination rules and the logic — are built by hand, and every page is read and checked by a person before it ships. Nothing is published on the strength of a model's claim about how a naming system works.
How do I report a mistake?
Email [email protected] with the page and what is wrong. Lore and linguistics corrections get priority, and we would rather be corrected than be confidently wrong in front of your gaming group.
Are the generators free to use commercially?
Yes. Names produced here carry no attribution requirement and no licence — use them in published fiction, video games, tabletop products, brands or usernames. We claim no rights over what the generator hands you.
Last reviewed . Questions about this policy: [email protected].