humanflow

Nominalization finder

Finds the places where a verb has been turned into a noun and propped up by an empty one. It flags the phrase and names the verb, rather than red-lining every word that happens to end in -tion.

Runs entirely in your browser — nothing is uploaded, and there is no account or limit.

Last reviewed 15 August 2026 · The HumanFlow team

What a buried verb costs

The committee made a decision to postpone the review. Nine words to say what the committee postponed the review says in five. The verb that carried the meaning — decide — has been packed into a noun, and made has been brought in to hold the sentence together while carrying nothing at all.

Doing this once is invisible. Doing it in every third sentence produces prose that feels heavy without any individual sentence being wrong, and it is one of the harder habits to see in your own draft, because each instance looks fine as you write it.

When the noun is the right choice

Often. If the process itself is your subject — comparing two methods of assessment, arguing that a classification is unstable — then naming it as a noun is precise, and the verb form would force you into a clumsier sentence. Writing guidance that treats every nominalization as an error is giving you a rule where it should be giving you a question: is this noun the thing I am talking about, or is it a verb I have flattened?

That is why this tool suggests rather than corrects, and why nothing here produces a score you can chase. The glossary entry for nominalization works through three sentences in detail, including one that is better left alone.

The habit it most often travels with is passive voice, and the two compound each other: “a decision was made” buries the verb and removes the person who made it in the same four words. They are separate mechanisms and separately fixable, which is why this tool does not flag passives — the reliable test for those is a sentence you can apply yourself, and that page has it.

Where it sits alongside the other tools

This one looks at grammar — a construction, not a vocabulary list. If you want the opposite angle, the AI word finder flags stock terms wherever they appear, and the sentence opener checker looks at how your sentences begin. For overall difficulty rather than any specific habit, the readability checker gives you a grade level with its working shown.

None of them is a detector and none of them predicts one. If that is the actual worry, how detectors work is the honest starting point, and stylometry covers what measuring style can and cannot establish about who wrote something.

Common questions

What is a nominalization?
A verb or adjective converted into a noun: decide becomes decision, apply becomes application, dense becomes density. The word is not a fault. The problem starts when the noun takes over the sentence and a near-empty verb has to be brought in to carry it — “conducted an investigation” doing the work that “investigated” did on its own.
Why does it not flag every word ending in -tion?
Because that tool would be useless. Information, education, nation, question and situation all carry the ending and none of them is a buried verb. A scanner working on suffixes alone red-lines an entire page of ordinary academic prose, which trains you to ignore it. This looks for the pattern — empty verb, then noun with a verb form available — and only reports a phrase when it can name the verb that would replace it.
Should I remove every one it finds?
No. Read each suggestion in its sentence and keep the ones that are doing something. Nominalizations let you make a process the subject of a sentence, which is often exactly what a methods section or a literature review needs, and rewriting mechanically towards verbs produces prose that is shorter and worse.
Is there a nominalization percentage I should aim for?
No, and be wary of any tool that gives you one. The density figure here is context, not a target — it varies enormously by discipline, and legal, scientific and administrative writing all sit legitimately higher than a personal essay does.
Will this change my AI detection score?
Do not use it for that. Detectors do not measure nominalizations, and cutting them is an editing decision that stands on its own merits. Writing in plainer, more direct sentences may well read differently to a statistical model, but that is a side effect and nobody can promise you a number.
Why do the two sections count different things?
They answer different questions. The flagged phrases are the actionable ones, found by pattern, where a specific verb is available. The count underneath is a broader sweep by word ending, discounting the common nouns that only look like nominalizations, and it exists to tell you whether the habit is widespread rather than to be fixed item by item.

Need more than a utility?

HumanFlow rewrites AI drafts to read naturally and scores them for AI, with the signals behind the number — free up to 1,500 words per scan.

Try the humanizer free

Other free tools

Text similarity checkerLexical diversity calculatorReference checkerQuote checkerAI disclosure generatorSentence opener checkerInvisible character detectorWord & character counterCase converterText cleanerSentence rhythm analyzerText diff checkerWord frequency counterCharacter counter with platform limitsAI word finderEssay structure checkerReadability checkerPrompt library

← All free tools