Count letters, digits, and special characters with frequency analysis. Free online letter counter runs in your browser. Free, private, in-browser.
A letter counter analyzes text to show how many letters, digits, spaces, and special characters it contains. Beyond simple counting, it provides a frequency breakdown showing how often each letter appears — useful for cryptography, linguistics, writing analysis, and educational purposes.
In English text, the most common letters are E, T, A, O, I, N, S, H, R in that order. This tool shows you the actual frequency distribution of your specific text, which you can compare against expected patterns. This is the foundation of frequency analysis in cryptography.
The tool categorizes every character into four types: Letters (A-Z), Digits (0-9), Spaces (including tabs and newlines), and Special Characters (punctuation, symbols, and non-Latin characters). This gives you a complete picture of text composition.
In a large sample of ordinary English text, letters appear in a stable order: E near 12.7%, T near 9.1%, A 8.2%, O 7.5%, I 7.0%, N 6.7%, S 6.3%, H 6.1%, R 6.0%, then D, L, C, U and the rest trailing down to J, Q, X and Z below 0.2% each. Your own text will not match those numbers exactly — a short passage, a technical vocabulary, or a name repeated many times shifts the distribution. As a rule of thumb, samples under about 200 letters are too small to compare against the reference figures with any confidence, while a few thousand letters usually settle close to them.
Deviations are informative rather than wrong. Legal text pushes T and H up because of "the" and "that". Code and configuration text raises E and N while collapsing the vowel balance. A passage full of Spanish or French loan words lifts the counts for accented characters, which this tool groups under special characters rather than folding them into their unaccented base letters.
A simple substitution cipher replaces each letter with another letter consistently, which hides the alphabet but preserves the statistics. That is its fatal weakness. Count the letters in the ciphertext, sort them, and the most frequent symbol is very likely to stand for E, with the next few covering T, A, O, I and N in some order. Confirm the guess with structure rather than counts alone: one-letter words are A or I, the most common three-letter word is almost always "the", and a doubled letter at the end of a short word often marks LL, SS, EE or OO.
Caesar shifts are even easier because every letter moves by the same amount — once you know the most frequent ciphertext letter, the shift is simply its distance from E. Polyalphabetic ciphers such as Vigenère defeat a single frequency table because each position uses a different alphabet; there you must first estimate the key length, then run a separate frequency count on every nth letter. This tool supports that workflow directly: paste each extracted sequence in turn and compare its distribution to the reference figures above.
Letter frequency is one of the few cryptography topics that can be demonstrated end to end without software or mathematics. A class can encipher a short message by hand, exchange messages, then break them using nothing but a frequency table and patience. The exercise makes an abstract security principle concrete: a cipher that leaks statistics leaks the message, which is exactly why modern encryption is designed so that output frequencies look uniform.
The same data supports language work. Comparing the letter profile of two authors, or of a translated passage against its original, shows measurable differences in spelling conventions and vocabulary. It is also a practical way to explain why Scrabble sets and typing layouts distribute letters unevenly — E has twelve tiles and Q has one for the same reason it dominates this table.
The special-character bucket is where surprises hide. Non-breaking spaces copied out of a word processor, curly quotation marks, zero-width joiners inside emoji, and invisible direction marks from right-to-left text all count as special characters and all inflate a total length that looks too high for the visible text. If a form rejects a message that appears short enough, checking the special-character count is the fastest way to find the cause.
The totals also matter for platform limits. A single SMS segment holds 160 GSM characters, but drops to 70 as soon as one emoji or non-Latin character appears. A post on X allows 280. A search-result title is safe to about 60 characters and a meta description to roughly 155. Those budgets are measured in characters, not words, so the total on this page is the figure to watch when you are writing into a fixed slot.
All analysis runs entirely in your browser. Your text is never uploaded or stored anywhere, which matters when the text is ciphertext, a password candidate, or unpublished writing. You can confirm it by watching your browser's network panel while you type — no request leaves the page.
Need a full word count? Try our Word Counter. Want to count sentences instead? Use the Sentence Counter. Compare length targets with the Text Length Analyzer or check clarity using the Readability Score Checker.
Only A-Z characters (case-insensitive) are counted as letters. Accented characters like é are counted as special characters.
Letters are sorted by frequency (most common first). Ties are in alphabetical order.
The frequency table tracks A-Z Latin letters. Characters from other scripts are counted under 'Special Characters'.
No hard limit. The tool runs in your browser and handles large texts efficiently.
Browse all free tools · Guides and tutorials · PDF tools · Developer tools · Text tools · SEO tools