GUIDE · CASE
Choosing between sentence case and Title Case
Letter case is a formatting decision with real conventions behind it. Pick the wrong one and a heading looks casual where it should look formal, or pompous where it should read plainly. This guide covers what each style is for, exactly what the converter does to your text, and the cases where no automatic conversion can be trusted without a human read.
The four styles and their jobs
- Sentence case — capitalize the first letter and proper nouns only. Used for body text, table cells, button labels and most sentences. It reads like ordinary writing.
- Title Case — capitalize most or all principal words. Used for book titles, article headlines, section headings and form titles. It scans well at a glance.
- UPPERCASE — used sparingly for short labels: badges, table headers, footers, signage. In long text it reads as shouting.
- lowercase — intentional in some brands and informal contexts, and the safe intermediate when you are normalizing mixed input before applying another style.
What each action actually does
It helps to know the exact rule, because the interesting failures come from the rule rather than from a bug. Here is what each button applies:
| Action | Rule applied | Example |
|---|---|---|
| Sentence case | Lowercase everything, then capitalize the first letter of the text and the first letter after a full stop, exclamation mark or question mark followed by whitespace | “HELLO WORLD. SECOND LINE” → “Hello world. Second line” |
| Title Case | Lowercase everything, then capitalize the first letter of every word, where a word starts at the beginning of the text or after any character that is not a letter or a digit | “the QUICK brown-fox” → “The Quick Brown-Fox” |
| UPPERCASE | Convert every letter to its uppercase form | “MiXeD” → “MIXED” |
| lowercase | Convert every letter to its lowercase form | “MiXeD” → “mixed” |
Worked examples
The same messy input, run through each action in turn:
input: the QUICK brown-fox JUMPS over 2 lazy DOGS sentence: The quick brown-fox jumps over 2 lazy dogs title: The Quick Brown-Fox Jumps Over 2 Lazy Dogs upper: THE QUICK BROWN-FOX JUMPS OVER 2 LAZY DOGS lower: the quick brown-fox jumps over 2 lazy dogs
Notice what Title Case did to brown-fox: the hyphen is not a letter, so the letter after it was treated as the start of a new word and capitalized. That is the mechanical rule, and it is also what most style guides ask for. Notice too that the digits are untouched and that over was capitalized even though many guides would leave a short preposition lowercase.
Why Title Case is never fully automatic
English style guides disagree about which words to capitalize in a title. Most agree that articles, short prepositions and coordinating conjunctions stay lowercase — a, the, in, of, and — but “short” means different lengths in different guides, and the same word changes role: through is a preposition in “walking through the park” and an adverb in “follow through”. A converter applies one consistent rule and cannot inspect meaning, so it will always produce a defensible starting point rather than a finished headline. The style-guide comparison goes through the differences in detail.
Names, acronyms and specialist terms add a second layer. Conversion cannot invent capitalization for words that should keep it in the middle — iPhone, van Gogh, unit symbols such as kg — and lowercasing destroys capitalization that carries meaning, so the acronym RAM becomes the animal ram. Neither outcome is a bug; it is the limit of formatting without context.
Words that conversion cannot protect
- Brand and product names with internal capitals: iPhone, eBay, YouTube, JavaScript.
- Acronyms that would otherwise read as ordinary words: RAM, NATO, LASER.
- Personal names with particles or unusual casing: van Gogh, de la Cruz, McDonald.
- Scientific and unit notation where case is part of the symbol: kg, mL, Hz, Pa.
- Species and genus names, where only the genus is capitalized.
- Non-Latin scripts with no case at all — case actions leave them unchanged, which is correct but can look like nothing happened.
A practical workflow
- Normalize first. Run lowercase over mixed input so every word starts from the same state and you can see the real content.
- Apply the target style. Sentence case for prose, Title Case for headings, titles and short labels.
- Restore what the rule cannot know. Fix brand names, re-capitalize acronyms, and adjust the short function words against the style guide you follow.
- Keep one style per block. A list of headings reads far better with consistent casing than with per-item choices.
- Read it aloud once. Case errors are almost always audible before they are visible.
Which style to choose
If the text will be read as a sentence or a paragraph, use sentence case. If it is a heading, a title or a short label that people scan rather than read, use Title Case. If you are normalizing data before a later step, use lowercase as the intermediate. The fastest test is to read the line the way it will actually be read: silently and fast for a headline, carefully and word by word for a sentence. Whichever you pick, apply it consistently — inconsistency between two adjacent headings is more noticeable than either style on its own.
Frequently asked questions
Why does sentence case turn my second line lowercase?
Because only a full stop, exclamation mark or question mark followed by a space starts a new sentence. A line break is not treated as a sentence boundary, so text after a wrapped line is lowercased and not re-capitalized.
Why did Title Case capitalize the word after a hyphen?
Because the rule treats the start of a word as the beginning of the text or any position following a character that is not a letter or a digit. A hyphen is not a letter, so the next letter counts as a new word.
Can I convert only part of the text?
The actions apply to the whole input panel. To convert one paragraph, put that paragraph in the input, transform it, and copy the result out before loading the rest.
Does case conversion work for languages without upper and lower case?
It leaves them unchanged. Scripts such as Chinese, Japanese, Korean, Arabic and Hebrew have no case distinction, so the action has nothing to change — which is the correct behavior rather than an error.