Letter Counter
Count letter frequency
Count letter frequency Runs locally in your browser
Counts a-z only, case-insensitive, and reports each letter as a share of total characters.
About the
The Letter Counter counts how often each letter appears in a piece of text and reports each one as a share of the total characters. It is the fastest way to spot an encoding problem, a suspiciously even distribution, or a text that has been through a substitution cipher, since English has a very distinctive letter profile. Counts are case-insensitive and cover a to z only, which keeps the report to what you can act on.
Where people use the
- Checking whether a text is what it claims to be by its letter profile.
- Starting a frequency analysis on a substitution cipher.
- Spotting a file that has been mangled by the wrong encoding.
- Settling an argument about which letter is most common in a document.
How to use the
- 1
Paste the text
Any length works, from a single sentence to a whole document.
- 2
Run it
Each letter from a to z is listed with its count and its percentage, ordered from most to least frequent.
- 3
Compare against English
In ordinary English prose, e, t, a, o and i lead the table. If they do not, the text is probably not plain English.
- 4
Use it as evidence, not proof
A frequency profile narrows things down. It does not decode anything for you.
Tips and best practices
- Compare the top of the list against e, t, a, o, i to sanity-check English text.
- Accented letters are not counted; normalise the text first if that matters.
- Percentages are shares of all characters, so they will not total 100.
Frequently asked questions
Are uppercase and lowercase counted together?
Yes. The text is lowercased before counting, so A and a are the same letter.
What about punctuation, digits and spaces?
They are excluded from the per-letter counts. They are included in the total used for the percentages, which is why the shares may not add to 100 percent.
Are accented letters counted?
No. Only the unaccented letters a to z are counted, so e and é are treated separately and only the first appears.
Why do the percentages not add up to 100?
Because the total includes spaces, punctuation and digits, which are not listed as letters.
Can this break a cipher?
Not by itself. Frequency analysis is the starting point for a substitution cipher, but you still have to do the mapping.