Formalizer

What academic register actually is

“Write more formally” is the least useful feedback a marker gives, because it names a problem without naming a cause. In practice, informal academic writing is almost always one of six specific habits, and every one of them is visible in the text rather than a matter of taste.

That is what makes this checkable by a tool at all. None of the six requires understanding your argument — they are patterns, and patterns can be found and replaced. Anything that does require understanding your argument is out of scope here, deliberately.

The six things it looks for

1. Contractions

Doesn’t, can’t, it’s, there’s. Contractions belong to speech. Academic prose writes them out, and this is the single most common reason a first-year essay reads as casual. It is also the easiest thing on this list to fix, which is why it is worth doing first.

One trap the tool avoids: the student’s essay is a possessive, not a contraction, and turning it into the student is essay is exactly the kind of silent damage a careless find-and-replace does.

2. Informal vocabulary

A lot of, stuff, things, big, really, basically. These are not wrong words; they are imprecise ones. A lot of evidence tells a reader nothing that considerable evidence does not, and the second sounds like someone who has read the literature.

Intensifiers are the subcategory worth being ruthless about. Very, really and extremely add emphasis but no information, and in academic writing emphasis without information reads as padding.

3. First person

I think, I believe, in my opinion. This is the one flag you should read rather than accept automatically, because the rule genuinely differs by discipline.

Most sciences and social sciences still expect claims to be made impersonally: not I think the data shows X but the data indicates X. A growing number of humanities courses positively encourage the first person, on the grounds that hiding the author is a pretence. Your module handbook decides. What is universal is that I think in front of a claim you have evidence for weakens it — you are telling the reader it is an opinion when you could be telling them it is a finding.

4. Hedging

It could be argued that, perhaps, somewhat, rather, arguably. Hedging is the most interesting item here because it is sometimes exactly right and sometimes a failure of nerve.

If your evidence is genuinely partial, saying so is honest and good scholarship. But it could be argued that is usually a way of making a claim while declining to own it, and markers read it as such. The test: can you say what the evidence actually supports? If yes, say that instead. If no, the problem is the evidence, not the phrasing.

5. Wordy phrases

Due to the fact that, in order to, at this point in time. Four words doing the work of one. These accumulate quietly — a 2,000-word essay commonly carries 150 words of pure connective padding, which is 7.5% of a word limit you probably need.

Cutting these is the fastest way to bring an over-length essay down without losing any content at all. If you are over the limit, start here rather than deleting a paragraph.

6. Two-word verbs

Find out, look into, come up with, carry out. Phrasal verbs are ordinary English and there is nothing wrong with them, but academic writing conventionally prefers the single-word Latinate form: determine, investigate, devise, conduct.

This is the most arguable of the six. Plain-English guidance often says the opposite, and for public-facing writing it is right. For an essay being marked against academic conventions, the single-word form is the safer choice.

Why it does not rewrite for you

Tools that rewrite a paragraph into academic register exist, and they need a language model to do it. Three reasons this one does not.

It would have to send your work somewhere. Every tool on this site runs in your browser precisely so that unsubmitted coursework never leaves your machine. Rewriting means uploading, and that is a promise we are not willing to break for a feature.

It would sometimes change your meaning. A rewriter that is right 95% of the time is wrong once per essay, and a sentence that now argues something slightly different from what you meant is much worse than a sentence that reads informally. You would not notice, because it would still read well.

It edges towards work that is not yours. Where the line sits between proofreading and having something written for you is a question your institution has an answer to, and a tool that replaces your sentences with its own is on the wrong side of most of them. Flagging a contraction is not.

How to use the score

The number is flagged items per hundred words, weighted by how much each kind matters, turned into a 0–100 figure. It is deliberately crude.

It is useful for one thing: comparing two drafts of your own writing. A first draft at 62 and a second at 88 tells you the editing worked. Comparing your score to someone else’s, or to a target, tells you nothing — a philosophy essay and a lab report have legitimately different registers, and no marker has ever awarded points for a formality score.

Above 90 is academic register. Below 55 usually means the piece was written the way you would explain it out loud, which is a perfectly good first draft and a poor final one.

Where this sits with the other tools

The readability checker answers a different question: not whether your writing is formal, but whether it is followable — sentence length, syllable density, passive constructions. A piece can be perfectly formal and still unreadable, and the two tools disagree often enough to be worth running both.

If you are cutting to a word limit, the character and word counter gives you the figure your marker will see, and the wordy phrases flagged here are the cheapest words to lose. And once the prose is right, the citation generator handles the references.