What this site does not publish, and why

NewbornNames publishes one kind of fact: how many U.S. babies of one sex received an exact spelling in one year, and where that count ranks. Everything on the site — the generator, the top-100 tables, the 52 letter pages, these guides — is computed from three files derived from the U.S. Social Security Administration's national baby-names dataset: the top 1,000 names per sex for 2025, the top 50 per starting letter per sex for 2025, and the top 200 per sex for 7 snapshot years (1955, 1965, 1975, 1985, 1995, 2005, 2015). Those files bound what the site can say, and this guide walks the boundary: each omission, its provenance reason, and what the data offers in its place.

The reasons are not editorial taste. The methodology page documents a single pipeline that selects subsets of the SSA archive and attaches ranks; it never edits a name or a count, and it never consults a second source. Anything that cannot be computed from those files has no checkable origin on this site, so it is not published.

No name meanings, origins, or etymologies

The site attaches no meaning, origin, language, or etymology to any name, including the 2025 leaders: Olivia, ranked #1 for girls with 13,544 registered births, and Liam, #1 for boys with 20,818. The provenance record states the reason in one line: there is no provenance-clean source for meanings, so v1 publishes none rather than fabricating or scraping any. The SSA files carry three fields per line — name, sex, count — and a meaning attached to Olivia would have to come from somewhere else.

What the site offers instead is the data the files do hold, read at a finer grain: every letter page is a computed profile of one initial. For girls in 2025, the initial with the largest share of births inside the top 1,000 is A: 170 of those 1,000 names begin with it, together taking 16.7% of the births, led by Amelia at #4 with 12,699. Of the 50 names on the A list for girls, 8 are short (four letters or fewer on the SSA spelling), 39 are medium (five to seven), and 3 are long (eight or more). For boys the leading initial in 2025 is J, with 91 names and 11.2% of top-1,000 births, led by James (11,945) on the J list for boys. None of that says what a name means; it says how often each was registered.

No "unique" or "rare" scores

The site never scores how uncommon a name is beyond stating its rank and count, and it attaches the SSA's privacy rule every time: names given fewer than 5 times to babies of one sex in a year are excluded from the files entirely. A name that is absent is not shown to be unused — it was registered fewer than 5 times, or not at all, and the data cannot say which. A uniqueness score would have to pretend otherwise.

The committed depth adds a second limit. The 2025 list for girls stops at rank #1,000, Harmoni, with 252 births; for boys it stops at #1,000, Langston, with 227. Below those ranks the site holds nothing, so it says "outside the committed depth" rather than guessing. The per-letter lists reach past rank 1,000 for some initials, because they take the top 50 of each letter regardless of overall rank: on the 2025 boys' U list, 48 of its 50 names rank past #1,000, down to Uzay at #13,746 with 5 births. 3 of the 52 letter lists are shorter than 50 because the 2025 file holds fewer qualifying names for the initial — the girls' Q list with 34, the girls' U list with 32 and the boys' Q list with 39 — and nothing is padded in to fill them.

No AI-generated names

The generator on the homepage does not generate names. It filters one of the committed lists — the 2025 top 1,000 for a sex, or a snapshot's top 200 — by starting letter, length class, and popularity tier, then draws a uniform random sample from whatever remains, using the browser's cryptographic byte source and rejection sampling. The sampler can only return positions in that filtered list, so every name it shows is a row from the same table, with the same rank and count.

That is checkable at build time. The sampling function accepts an injected byte source; feeding it a fixed byte sequence instead of the browser's random bytes makes the draw identical on every build. Over the 2025 boys' list of 1,000 names, that fixed draw returns 3 entries: Noah (#2, 20,358 births), Marcos (#517, 579 births) and Mateo (#8, 11,045 births). Each is a real 2025 registration row, not a blend or a spelling the files do not contain.

Pools are sized by the data and never padded. There are 3 short girl names beginning with A inside the 2025 top 100, so that is the most the generator can draw for that filter. Switch the era to "2010s (2015 snapshot)" and the list is 200 names deep; the tier covering ranks 501–1,000 begins past that depth, so the generator reports an empty result rather than substituting names, and the tier for ranks 101–500 holds only 100 girl names in the 2015 snapshot. An empty result is a real result.

No merged spellings

The SSA counts each spelling as its own entry, and so does the site. The SSA's background page uses Caitlin, Kaitlyn, and Katelyn as its example; the provenance record restates it and adds that a "true" popularity merging spellings would differ and is not computed here. Looked up in the 2025 per-letter girls' lists, Kaitlyn ranks #802 with 340 births, while Caitlin and Katelyn fall outside the committed top 50 for their letters. A merged figure would be the site's number, not the SSA's, so the rows stay separate.

Two further SSA rules shape what a "spelling" is. Hyphens and spaces are removed from the first-name field before counting, so Julie-Anne, Julie Anne, and Julieanne are one entry, and the site cannot separate them. And rankings are per sex: a name on both lists has a separate rank and count on each. On the 2025 O lists, Oakley is #156 for girls with 1,933 births and #454 for boys with 688; the site reports both figures and never adds them.

No years or places outside the committed files

The SSA archive holds one file per birth year from 1880 to 2025, and the site commits 8 of them: 2025 at a depth of 1,000 per sex, and the snapshot years 1955, 1965, 1975, 1985, 1995, 2005, 2015 at a depth of 200 per sex. A decade era in the generator is one of those single years — "1950s" means registered in 1955 — and no figure for any other year appears on the site; nothing is interpolated between snapshots.

The place limit is the SSA's own: national data covers births in the 50 states and the District of Columbia, births in U.S. territories are tabulated separately and not included, and the data says nothing about name popularity outside the United States.

What the snapshots do support is a leader per sex per year, read directly from each committed file:

They also support a per-initial share of each snapshot's top-200 births. Between the 1955 and 2015 snapshots, the initial whose share of top-200 girl births increased most is A, from 3.0% to 20.2%; the largest decrease is D, from 13.8% to 0.8%. The site states the two figures and stops: the snapshots can show that a share changed, not why.

No labels, scores, or advice

The about page puts it plainly: popularity is a fact, and whether a name is right for a child is not something a dataset can settle. The site never scores, recommends, or ranks names by quality, and a low count is not evidence a name is strange, nor a high one that it is safe. The same applies to direction words. A name's rank in the 2015 snapshot and its rank in 2025 are two facts the files hold, and the site prints both rather than summarising them in an adjective.

What the site publishes instead

• A rank and a count for every name shown. The top 100 girl names and top 100 boy names of 2025 open at Olivia (13,544) and Liam (20,818).
• A computed profile for each of the 26 initials per sex: its share of top-1,000 births in 2025, its split by length class and popularity tier, the names on both sexes' lists, and its presence in each snapshot's top 200.
• 7 decade snapshots from 1955 to 2015, each 200 names deep per sex and labeled with the single year it comes from.
• A generator that samples those lists uniformly and reports an empty pool as empty.

Frequently asked questions

Related pages

The methodology page documents the pipeline and the SSA's qualifications in full; the about page explains why the site holds to one source; the guides index lists the other data guides.

Source: every name, rank, and count on this page comes from the U.S. Social Security Administration's national baby-names dataset (1955–2025 data, from Social Security card applications). It is public-domain U.S. government data; names given fewer than 5 times in a year are excluded by the SSA, spelling variants are counted separately, and the data covers U.S. births only. Full caveats on the methodology page.