Passive Voice & Grammar Checker

Clear, active writing engages readers, strengthens arguments, and boosts conversions. This free, rule-based checker detects passive voice constructions, grammatical agreement errors, confusable homophones, and wordy fluff in real time. Everything runs directly in your browser, keeping your drafts completely private and secure.

What is a rule-based passive voice and grammar checker?

A rule-based passive voice and grammar checker is an editorial utility that audits prose against established grammatical principles, morphological patterns, and syntactic heuristics. Unlike statistical or generative artificial intelligence engines that guess probabilities based on large language models, a deterministic rule-based engine tests text against explicit linguistic criteria: auxiliary verb chains, past participle inflections, phonetic article agreements, confusable word pairs, and known redundant phrases.

The primary advantage of a client-side, rule-based approach is predictability and complete privacy. Every character typed into this tool is parsed in memory inside your web browser using JavaScript. No draft is transmitted to an API, stored in a cloud database, or used to train third-party machine learning models. Whether you are drafting a confidential cold email outreach sequence, optimizing high-stakes SaaS sales copy, or polishing an academic research abstract, your intellectual property never leaves your machine.

How to use this passive voice and grammar checker

1. Paste or type your draft into the main editing area. You can check anything from a single sentence to an entire blog article or sales page.
2. Choose your writing style and target: Select Direct Web & SaaS for aggressive active voice optimization (aiming for under 10% passive voice), Standard Editorial for balanced articles, or Formal & Academic when objective technical phrasing is expected.
3. Examine the quality score and health meters: The central status gauge calculates an overall rating from 0 to 100 alongside category breakdowns for voice directness, grammatical mechanics, concision, and punctuation hygiene.
4. Review marked tokens in the interactive inspector: The visual text box annotates every flagged phrase with color-coded underlines, making it easy to identify where passive constructions or agreement errors occur in your paragraphs.
5. Filter diagnostics and apply fixes: Use the category filter tabs (Passive, Grammar, Wordy, Punctuation) to focus on one editorial dimension at a time. Each finding highlights the offending phrase and suggests a direct rewrite.
6. Copy your text or export a report: Use Copy text to transfer your draft back to your CMS or email client, or click Copy report to generate a formatted markdown summary for peer review or client sign-off.

How the quality score is calculated

The overall writing quality score is a weighted blend of four independent categorical metrics, calibrated on a 0 to 100 scale:

Category	Weight	What it measures	Ideal benchmark
Voice & Directness	35%	The percentage of sentences containing passive voice constructions (auxiliary “to be” + past participle).	<10% for SaaS & web copy; <15% for general editorial.
Grammar & Mechanics	35%	Subject-verb agreement, phonetic article matching (a vs. an), auxiliary verb forms (e.g., could have vs. could of), duplicate words, and confusable homophones.	0 errors. Deductions scale with text length to prevent single-typo distortion on long manuscripts.
Concision & Flow	20%	Detection of bureaucratic filler phrases (e.g., “in order to”, “due to the fact that”, “utilize”) and redundant modifiers (e.g., “past history”, “end result”).	0 dead-weight phrases. Plain, punchy phrasing is rewarded.
Punctuation & Hygiene	10%	Spurious whitespace before punctuation marks, duplicate commas or periods, and spacing consistency.	Clean typographical boundaries without accidental spacing glitches.

Scores map to standard editorial letter grades: Grade A (90–100) indicates clean, publish-ready copy; Grade B (75–89) indicates strong writing with minor phrasing to polish; Grade C (60–74) highlights frequent passive voice or wordy habits; Grade D (45–59) indicates heavy passive construction or grammatical clashes; and Grade F (0–44) flags severe mechanical issues requiring revision.

Anatomy of passive voice in English

Understanding what passive voice actually is—and what it is not—is essential for every writer. Passive voice is a grammatical voice where the grammatical subject of the sentence receives the action rather than performing it.

Every genuine passive voice construction requires two core elements:

1. A form of the auxiliary verb “to be” (or “get”): am, is, are, was, were, be, been, being, get, got, gotten.
2. A past participle (V3) of the main verb: either a regular verb ending in -ed (such as launched, tested, developed) or an irregular past participle (such as written, built, chosen, sent).

Consider the contrasting syntax:

• Passive: “The conversion rate was increased by our copywriter.” (The subject “conversion rate” receives the action; the actor “copywriter” is relegated to a prepositional phrase).
• Active: “Our copywriter increased the conversion rate.” (The actor “copywriter” directly executes the verb “increased” upon the direct object).

Furthermore, passive constructions frequently omit the actor entirely, creating what linguists call an agentless passive: “A mistake was made” or “The database has been updated.” While agentless passives are common in technical release notes, they often weaken marketing and sales copy by obscuring who is responsible for the result.

Common false positives that are not passive voice

Many amateur writers mistakenly equate any appearance of “is” or “was” with passive voice. The following constructions are grammatically active, and this tool specifically distinguishes them:

• Active Progressive (Continuous Tense): “The marketing team is testing a new headline.” This combines “is” with a present participle (-ing). It is active voice because the team is actively performing the testing.
• Predicate Adjectives: “Our founder is tired” or “The file is closed.” Here, “tired” and “closed” describe a state or attribute rather than an incoming action. This tool maintains an exclusion lexicon for common predicate adjectives (such as interested, excited, delighted, married, prepared) to avoid false warnings.
• Linking Verbs: “The pricing tier is affordable.” The verb “is” simply acts as a copula connecting the subject to its complement.

Why passive voice weakens copywriting and sales conversion

In copywriting, clarity and velocity determine whether a visitor converts or bounces. Passive voice degrades copy performance across three critical areas:

1. Diluted Energy and Momentum: Active sentences front-load the agent and verb, propelling the reader forward. Passive sentences delay the action, making value propositions feel sluggish and bureaucratic. Compare “Sign up today and get automated reports generated for your team” with “Sign up today and Antigravity generates automated reports for your team.”
2. Missing Accountability: In customer support and product messaging, passive constructions sound evasive. Writing “Your account was suspended because rules were violated” feels cold and unhelpful compared to “We paused your account because your sending volume exceeded our safety threshold.”
3. Word Bloat: Passive constructions almost always require extra auxiliary verbs and prepositions (such as “has been configured by” vs. “configured”). In email subject lines, mobile hero sections, and ad copy where character limits matter, passive voice wastes valuable screen real estate.

When passive voice is actually preferred

Active voice is not a dogma; it is an editorial default. Professional editors recognize specific situations where passive voice is rhetorically superior:

• When the actor is unknown or irrelevant: “The package was delivered at 9:00 AM.” The recipient cares about the arrival of the package, not the name of the courier driver.
• When the recipient of the action is the true topic: “President Kennedy was inaugurated in 1961.” Kennedy is the focus of the biographical paragraph, not the Chief Justice administering the oath.
• In scientific and technical methodologies: “The solution was heated to 85°C for three hours.” The procedural condition must be repeatable regardless of which laboratory assistant held the beaker.
• To soften bad news or avoid finger-pointing: “An error was encountered during checkout” is gentler than “You entered an invalid billing address.”

For this reason, this tool provides a Writing style & target selector. Choosing Formal & Academic broadens the tolerance band, acknowledging that technical and academic manuscripts legitimately require higher passive voice density than high-converting SaaS landing pages.

What each rule category checks for

1. Passive voice constructions

The parser tokenizes the text into syntactic word spans, identifies forms of “to be” or “get”, accommodates up to three intervening adverbs (such as “was recently and thoroughly tested”), and inspects the following verb against a comprehensive index of irregular past participles and regular -ed verbal stems. When followed by an optional “by [agent]” phrase, the tool extracts the agent and suggests an inverted active sentence structure.

2. Indefinite article agreement (a vs. an)

The choice between “a” and “an” in English is governed by phonetics (sound), not spelling. Many software checkers make the error of simply checking whether the following letter is a vowel. This tool checks phonetic pronunciation patterns:

• It flags “a apple” or “a error”, but correctly permits “a user”, “a European”, and “a unique feature” because these words begin with the consonant glide /j/ (y-sound).
• It flags “an user” or “an university” as phonetic mismatches.
• It correctly demands “an hour” and “an honest review” because the leading “h” is silent.

3. Modal auxiliaries and colloquial traps

A frequent error in informal and conversational writing is substituting “of” for “have” in modal constructions (such as “could of”, “should of”, “would of”, “might of”, “must of”). This arises because the phonetic contraction “could’ve” sounds identical to “could of”. In formal and commercial writing, this is an egregious grammatical error that damages credibility.

4. Homophones and confusable word pairs

Contextual rules identify commonly confused word pairs that slip past traditional spellcheckers because both words exist in the dictionary:

• Their vs. There vs. They’re: Flags “their is” or “their are” (demanding existential “there”), and flags “they’re team” (demanding possessive “their”).
• Your vs. You’re: Flags “your welcome” or “your going to” (demanding “you’re”), and flags “you’re account” (demanding possessive “your”).
• Its vs. It’s: Flags “its a” or “its not” (demanding the contraction “it’s”), and flags “it’s features” (demanding the possessive pronoun “its” without an apostrophe).
• Then vs. Than: Flags comparative errors such as “faster then”, “better results then”, or “more leads then”, reminding the writer that “than” is required for comparisons.
• Loose vs. Lose: Flags “loose customers” or “loose revenue” where the verb “lose” is intended.
• Affect vs. Effect: Flags “take affect” (which requires the noun “effect”) and “the affect of”.
• Lead vs. Led: Flags “has lead to”, correcting the past participle to “led”.

5. Wordy filler and bureaucratic redundancies

Strong writing eliminates flab. The concision engine flags verbose idioms and supplies one-word plain English alternatives:

• “In order to” → “to”
• “Due to the fact that” → “because”
• “At this point in time” → “now”
• “For the purpose of” → “to”
• “In the event that” → “if”
• “Is able to” / “Has the ability to” → “can”
• “Utilize” → “use”
• Redundant word pairings: “each and every” (“every”), “first and foremost” (“first”), “future plans” (“plans”), “past history” (“history”), and “end result” (“result”).

6. Punctuation and typographical spacing

The punctuation analyzer flags typographical glitches that compromise polished presentation, including spaces before commas, periods, or colons (e.g., “word ,” or “sentence .”) and accidental double punctuation marks (such as “,,” or “..”).

Worked copywriting examples

Original Draft	Score & Grade	Key Issues Identified	Recommended Active Rewrite
“Our software was built by engineers to optimize onboarding. In order to utilize the API, an API key is required by the user.”	64 (Grade C)	2 passive constructions (“was built by”, “is required by”); 2 wordy phrases (“in order to”, “utilize”).	“Our engineers built this software to streamline onboarding. To use the API, generate your personal API key.”
“We should of followed up earlier. Their is a new feature that delivers higher reply rates then before.”	52 (Grade D)	Auxiliary error (“should of”); confusable homophone (“their is”); comparative mismatch (“rates then”).	“We should have followed up earlier. There is a new feature that delivers higher reply rates than before.”
“Antigravity automates outreach without burning your sending domain. Our system verifies email addresses in real time.”	100 (Grade A)	0 passive sentences; 0 grammatical errors; concise, punchy vocabulary.	None needed—clean, active benchmark.

Rule-based checkers vs. AI grammar tools: privacy and predictability

Large language model (LLM) grammar tools have become popular, but they introduce distinct tradeoffs that professional writers and business founders must weigh carefully:

• Data Privacy and Security: When you paste text into an AI grammar tool, your copy is sent across the internet to external server farms. If you work under nondisclosure agreements, handle medical or financial records, or write proprietary sales strategies, transmitting unreleased copy to cloud providers introduces regulatory and compliance risks. This tool processes text locally in your browser session; nothing is saved or logged.
• Predictability vs. Hallucination: Generative AI models frequently rewrite sentences to mean something different from what the author intended, smoothing away distinct brand voice or inventing facts. A deterministic rule-based checker flags exact syntactic violations without rewriting your authentic message.
• Zero Latency: Because rule-based regex parsing executes directly on the client machine, calculations occur in under 50 milliseconds without waiting for server network queues.

Limitations and accuracy

No client-side, rule-based checker can replace a human editor, and transparency regarding technical limitations is essential:

• Contextual Ambiguity: In English, some words serve as both past participles and adjectives depending on subtle semantic context. While our lexicon excludes dozens of common predicate adjectives (such as “experienced”, “tired”, “talented”), novel or ambiguous participles may occasionally be flagged. Treat every highlight as an editorial prompt to verify intent.
• Sentence Splitting Heuristics: The sentence tokenizer accounts for common abbreviations (such as e.g., i.e., vs., Dr., Inc.). However, unusual custom acronyms ending with periods followed by capitalized words may split a sentence prematurely.
• Dialect Variations: Rules prioritize standard international business English. Certain British, American, Australian, or Canadian idioms may have slight stylistic differences.

Frequently asked questions

What is the difference between active voice and passive voice?

In active voice, the subject of the sentence performs the action (“Sarah wrote the report”). In passive voice, the subject receives the action (“The report was written by Sarah”). Active voice is typically clearer, shorter, and more engaging for readers.

Is passive voice grammatically incorrect?

No. Passive voice is completely legitimate English grammar. It becomes an issue when writers overuse it unintentionally, which makes writing detached, wordy, or vague. In business writing, marketing, and web copy, keeping passive voice under 10% ensures energetic, persuasive communication.

Why does this tool flag “in order to” and “utilize”?

These phrases are classic examples of dead-weight filler. In almost every sentence, “in order to” can be replaced with “to” without changing the meaning, saving two words. Similarly, “utilize” is a multi-syllable substitute for “use” that adds unnecessary formality to plain English writing.

Does this tool store or upload my text?

No. The analyzer is 100% client-side. The JavaScript runs inside your browser sandbox. No text, analytics, or cookies are sent over the network, making it completely safe for proprietary drafts, confidential contracts, and client communications.

How does the phonetic “a” vs. “an” check work?

English rules state that “an” is used before vowel sounds, and “a” is used before consonant sounds. The checker incorporates phonetic exceptions, permitting “a user” (because “user” begins with a “y” consonant sound) and “an hour” (because the “h” is silent).

Can I export my diagnostic report?

Yes. Clicking the Copy report button compiles your overall score, category breakdown, word metrics, and a numbered list of all flagged issues into clean markdown format, ready to paste into GitHub, Jira, Notion, or Slack.

What does the “Writing style & target” selector do?

It adjusts the scoring curve for passive voice tolerance. Direct Web & SaaS mode expects under 10% passive voice to achieve an “A” grade. Standard Editorial allows up to 15%, while Formal & Academic accommodates up to 20% passive voice before applying score deductions.

How does this tool compare to Hemingway or Grammarly?

Like the Hemingway Editor, this tool highlights stylistic weaknesses like passive voice and wordiness with color-coded underlines. Unlike cloud-based grammar extensions, it requires no login, installs no background trackers, sends zero data to external servers, and costs nothing.