About this site

I build this site because the best data on American names is public, complete, and almost unusable. The Social Security Administration publishes every name given to five or more babies since 1880, plus the same broken out by state since 1910. It is a remarkable archive. It also arrives as a zip file of text files with no per-name pages, no living-population figures, and no way to ask the question most people actually have, which is not "was this name popular in 1974" but "how many people are called this, and how old are they".

So that is what this site computes. Every page starts from the source files and applies formulas written down in full on the methodology page. Where I have made a simplifying assumption — and the living-population estimate makes a large one — the page says so rather than presenting a number as more solid than it is. The same figures drive the year pages, the state pages and the data stories; none of it is written twice.

Why ninan.org

Ninan is a Malayalam given name and surname, carried mostly by Saint Thomas Christians in Kerala. It descends from Syriac Yohannan, which is the same name that reaches English as John by a different route. Wiktionary files the English entry as a borrowing from Malayalam.

The name does not appear in the SSA files at all. That is not because nobody in America is called Ninan; it is because in no single year did five or more American babies receive the name, so it never cleared the reporting floor. A site about name data sitting on a domain whose own name is invisible to the data is a decent reminder of what the floor hides: the tail of the distribution is enormous, and almost none of it is published anywhere.

The Indian-names section grew out of the same observation. Indian given names are being used in American and British births in numbers that are easy to measure and that nobody bothers to measure. The lists that dominate searches for Indian names are compiled from meaning dictionaries, not from birth records. This section uses the birth records.

Why the site grows slowly

There are 8,979 names that meet the bar for a page, and 600 published so far. The rest are released a few dozen a day, in an order that mixes the obvious names with less obvious ones, and the rate only increases while Google is actually indexing what has already gone out.

This is deliberate. A site that publishes thirty thousand pages in one afternoon is telling search engines exactly one thing about itself, and it is not a good thing. It also removes any chance to notice that a template is thin before it has been applied thirty thousand times. Publishing at a pace that can be checked is slower and produces a site that is worth checking.

Corrections

If a number here is wrong, I would rather know. Every metric can be recomputed from the source files using the formulas on the methodology page, and each name page links to a JSON version of its own data so you can check the arithmetic without scraping HTML. Get in touch.

Written and maintained by Alex Ninan. Data current to 2025.