How common is my name?
It turns out it's pretty common, actually. But the interesting part is how I know that.
11/18/2026

I love it when federal agencies open up their data for public consumption. Last year, for example, IBGE (Brazilian Institute of Geography and Statistics) published a site that compiles the frequency of first and last names based on the latest census, conducted in 2022. That tool (deservedly) got a lot of news attention and made it possible to answer the question that opens this post.
So, well, how common is my name? Luís is the 16th most common name in Brazil, while Fernando is the 38th. My surname, Bittencourt, ranks 516th. However, there's an interesting fact about Luís and Bittencourt: they're not the most common spellings. Luiz, with the Z, is the 10th most common name. Bitencourt, without a double T, ranks 444th. That got me thinking: how different would those names rank if we considered all their spellings?
I could answer that question by making a list of all spellings and searching the site one by one, but 1) that's not how software engineers behave and 2) I happened to have built a tool that does exactly that: String Expansion. In a nutshell, it expands a given pattern into all possible combinations. To get all spellings for my first and last names, for example, we can do:
import expand from 'string-expansion';
console.log(expand('Lui(s|z)'));
console.log(expand('Bitt?enc(+o|u)rt?'));The language was clearly inspired by regular expressions:
(s|z)generatessandz- A question mark
?makes what comes before (char or group) optional - A plus sign
+at the start of a group generates all group alternatives plus all of them concatenated. In the example above,(+o|u)generateso,uandou. We can achieve the same with(o|u|ou)but hey, we're engineers. Why not get fancy when we have the chance?
I wrote that package two years ago to solve the exact same problem (name spellings) in a different context. I came up with the grammar and implemented the first version using Chevrotain. Earlier this year, I used Claude Code to remove that dependency.
After all the nerd stuff, here are the results:
| Spelling | People | Rank |
|---|---|---|
| Luiz | 1,335,098 | 10th |
| Luis | 838,237 | 16th |
| Combined | 2,173,335 | 6th |
| Spelling | People | Rank |
|---|---|---|
| Fernando | 502,065 | 38th |
| Spelling | People | Rank |
|---|---|---|
| Bitencourt | 49,611 | 444th |
| Bittencourt | 38,067 | 516th |
| Bitencurt | 12,897 | 1,106th |
| Bitencur | 5,455 | 2,115th |
| Bittencurt | 3,814 | 2,811th |
| Bitencort | 2,139 | 4,598th |
| Bitencour | 1,947 | 4,985th |
| Bittencort | 817 | 11,082nd |
| Bittencour | 582 | 14,838th |
| Bittencur | 469 | 18,054th |
| Bitencor | 356 | 22,916th |
| Bittencor | 51 | 103,379th |
| Combined | 116,205 | 253rd |
As you can see, Luís jumps to the 6th position, while Bittencourt goes from 516th to 253rd. In other words, both are more common than one might think by looking at a single spelling. Also, it stands out that even the rarest spellings, such as Bittencor, have records. I probably wouldn't have thought of them if I hadn't used a combinatorial tool like String Expansion.
It's always nice to use coding skills to solve a problem, even if it only matters to me (or isn't a problem at all).