There is no public national directory that counts every first-and-last-name combination. A full-name result is therefore an estimate—but it can still be useful when the inputs and assumptions are explained honestly.
The two numbers we begin with
Our first-name and surname frequencies come from the U.S. Census Bureau’s 2020 names release. This matters because both inputs describe a national population snapshot. We do not treat the sum of 145 years of baby-name records as the number of people alive today.
Suppose the Census table contains 103,000 people named Maya and 1,085,838 people with the surname Rodriguez. Each count can be divided by the 2020 resident population to produce a national frequency.
First-name count × last-name count ÷ 2020 U.S. population
Using that formula, Maya Rodriguez produces an estimate of roughly 337 people. The number is best read as an order-of-magnitude answer: probably hundreds, rather than a handful or tens of thousands.
The independence assumption
The formula assumes that first and last names are distributed independently. In other words, it treats the chance of being named Maya as unrelated to the chance of having the surname Rodriguez.
Real names do not behave that neatly. Language, migration, religion, geography, and family tradition can make a first name more likely alongside some surnames than others. A Spanish given name and Spanish surname, for example, may co-occur more often than a purely national model predicts. Other combinations may occur less often.
This is the largest limitation of a national full-name estimate. The result is not a directory count, and it should never be used as evidence that a particular person exists.
Why we use the 2020 population denominator
The source counts describe the 2020 Census population, so the matching denominator is the official 2020 resident population of 331,449,281. Mixing a 2020 surname count with a later population estimate would make the calculation look more current without actually updating the underlying names data.
Publication thresholds affect rare names
The Census public name files include names occurring at least 100 times. If a first name or surname is absent, the correct conclusion is not “zero people.” The defensible conclusion is “fewer than 100 in the published national table.”
The SSA series has a separate privacy rule: a name-sex-year combination is omitted when fewer than five records occur. This means a rare name can have gaps in its historical timeline even when it was genuinely used.
How to interpret the answer
- Use the scale. The distinction between tens, hundreds, and thousands is more reliable than the final digit.
- Expect clustering. Regional and cultural communities will not resemble a perfectly random national sample.
- Treat alternate spellings separately. Sofia and Sophia are separate records, as are O’Neil and O’Neill.
- Do not infer identity. A name cannot determine a person’s race, sex, ancestry, religion, or nationality.
Why an estimate is still valuable
For ordinary curiosity, the useful question is often not whether the true number is exactly 337. It is whether the name is likely shared by two people, a few hundred people, or an entire stadium. A transparent model can answer that scale question while leaving room for the uncertainty the data cannot remove.