Live Writing Editor

Summary#

Live Writing Editor scores a draft as you type: Flesch reading ease, grade level, long sentences, complex words, passive constructions, adverbs and tone. It runs entirely in your browser, costs nothing, and has no run button.

Purpose#

You cannot read your own writing cold. You know what you meant, so the sentence that loses a stranger reads perfectly to you. Jargon creeps in unnoticed, and by the time bounce rate tells you a page is hard to read, the traffic has gone.

This tool exists to give you the outside eye you cannot give yourself, on every keystroke. The decision it supports is whether a sentence is ready — while you are still writing it, not after it ships.

Overview#

You type or paste into one large box. Roughly a fifth of a second after you stop typing, the analysis below the box redraws.

The maths is standard and published: Flesch Reading Ease and the Flesch-Kincaid grade level, computed from your average sentence length and average syllables per word. Around those, the editor counts sentences over 25 words, words of three or more syllables, passive constructions, and adverbs ending in "ly". It reads tone from positive and negative word counts, and reads formality from how often you use contractions and the words "you", "your", "we", "our" and "us".

Everything happens in your browser. Nothing is uploaded, nothing is stored on a server, and there is nothing to run. Below about five words the analysis area stays empty; from there it updates continuously.

Benefits#

• Free, with no credit cost and no plan gate.
• No run button — the score keeps up with your typing.
• Your draft never leaves the page, so unpublished work is safe to paste.
• Each weak check turns into a specific instruction, not a vague nudge.
• Reading ease and grade level are the same measures editors have used for decades, so the numbers mean something outside this product.

Use Cases#

Tightening landing copy. Watch reading ease climb as you cut corporate language out of a hero paragraph.

Writing to a target audience. Set a grade level you want and edit until you hit it, rather than guessing whether a draft has drifted too dense.

Editing help and documentation. Long sentences and passive voice are exactly what makes an instruction hard to follow, and both are flagged the moment they appear.

Reviewing someone else's draft. Paste it in and the five checks give you a concrete list to discuss instead of "make it clearer".

Writing in a second language. The checks name the exact sentence to shorten and the exact word to swap, in plain terms, with no editor to wait on.

Requirements#

• A signed-in account to reach the dashboard.
• A modern browser. No plan tier, no credits, no integration, no verified domain.
• At least five words of text before anything is scored.

Permissions#

Every plan tier can use it, including Free, and every team role sees the same screen. There is no reduced view for a lower tier — nothing here is gated, because nothing here calls the API.

Cost#

Free. 0 credits. All scoring runs locally in your browser, so no request is made, nothing is metered against your monthly quota, and nothing counts toward the hourly fair-use limit. See How credits work for what does cost credits.

Navigation Path#

Dashboard → Write & Optimize → Live Writing Editor

Inputs#

The hint line under the draft box reads Analysis updates automatically as you type. — there is no run control.

Step-by-Step Guide#

1. Open Write & Optimize and choose Live Writing Editor.
2. Type or paste your draft into the box.
3. Once you pass five words, the report appears below.
4. Read the four KPI tiles, then the Live checks list.
5. Work down Fix as you write from the top. The instructions are ordered by impact.
6. Edit in place and watch the numbers move. Keep going until Fix as you write is replaced by the all-clear.
7. Copy your finished text out of the box. The draft stays in your browser on this device, so you can also leave and pick it up later.

Reading the Results#

The four KPI tiles.

Reading ease. A meter with the score out of 100 and a color tier: green at 60 or above, amber from 30 to 59, red below 30. Higher is easier. Sixty is the plain-English zone most web readers expect; below 30 reads like a legal or academic paper. The score is driven by two things only — sentence length and word length — so the two fastest ways to move it are to split a sentence and to swap a long word for a short one.

Live checks. Five rows, each with its own tier badge:

The adverb threshold scales with length: a long piece is allowed more adverbs than a short one, with a floor of two.

Fix as you write. The section that turns the scores into instructions. It appears only when something needs attention, and each line is specific and counted, so you know when you are done:

• Reading ease is below plain-English. Shorten sentences and swap long words for simpler ones. — shown below 60.
• N long sentences. Break each into a single idea. — shown above 2.
• N% of words are 3+ syllables. Replace jargon with plainer alternatives where you can. — shown above 20%.
• N passive constructions. Rewrite as "X does Y" rather than "Y was done". — shown above 2.
• Reads formally. Add a few contractions and "you/we" to sound more human and on-brand. — shown on formal drafts over 60 words.

When nothing is flagged and you have at least 20 words, the section is replaced by a single line: Clear, readable, and on-brand. Keep going.

Reading the numbers together. Reading ease is the headline, but it is derived — it will not move until you change sentence length or word length. So work the checks, not the score: passive voice and complex words are the two fastest wins, because cutting them shortens sentences and words at the same time. Tone and formality are separate from readability. A draft can be perfectly readable and still sound like a memo; if Tone reads formal voice and your brand is warm, add contractions and address the reader directly.

What to do with it. Get reading ease to 60 or above, grade level to 8 or below, long sentences and passive voice to zero or one, and complex words under about 12%. At that point the draft is clear. Clear is not the same as citable — run GEO Content Scorer next if the piece needs to be quoted by AI answers.

Examples#

Screenshots#

Tips#

• Fix passive voice and complex words first. They move reading ease the most.
• Aim for grade 8 on general web writing. Technical or legal copy will sit higher, and that can be deliberate.
• One or two passive constructions is normal. The count exists to make the choice deliberate, not to eliminate the form.
• Paste a competitor's page to see what you are being read against.
• Work in sections on a long article. A single number across 3,000 words hides the one dense paragraph that is doing the damage.

Best Practices#

• Draft first, score second. Writing with the meter open from word one slows you down.
• Set a grade-level target per content type — lower for landing pages, higher for technical documentation — and edit to it.
• Read a red Long sentences count as a structure problem, not a punctuation problem: usually the sentence contains two ideas.
• Check Tone against your brand voice, not against green. Formal is correct for some brands.
• Run the readability pass before Pre-Publish Trust Checks, so you are fact-checking the sentences you intend to keep.

Common Mistakes#

• Chasing 100 reading ease. Above about 90 the writing starts to read as simplistic. Sixty to eighty is the target for most business writing.
• Eliminating every passive construction. Some sentences genuinely need one.
• Treating grade level as a quality score. It measures accessibility, not insight.
• Scoring a whole article at once and missing the paragraph that is dragging the average down.
• Expecting it to rewrite for you. It tells you what to change; you make the change.

Limitations#

• English only. The syllable counting and word lists are English rules, so scores on other languages are not meaningful.
• Passive voice is matched by pattern, so unusual constructions can be missed and the occasional false positive appears.
• Tone is a keyword tally over a fixed list of positive and negative words, not a semantic read. It has no opinion on sarcasm or context.
• Formality is judged from contractions and second-person words only.
• The analysis is live-only. It is not written to the Library and there are no export buttons on this screen — it rebuilds instantly from your draft, which stays in the box on this device.
• No spelling or grammar checking. This measures readability, not correctness.
• There is no Get recommendations panel on this tool.
• Markdown and HTML in the box are counted as words, so paste plain text for an accurate read.

Troubleshooting#

FAQs#

Where do the scores come from? Published readability formulas — Flesch Reading Ease and the Flesch-Kincaid grade level — plus counts of passive constructions, three-or-more-syllable words, "ly" adverbs, and positive and negative tone words. All of it runs on your device.

Does my text get sent anywhere? No. Every score is computed in your browser as you type. Your draft never leaves the page, so unpublished work is safe to paste. See What data we store.

What counts as a good reading ease score? Sixty or higher for most web writing, which lands around an eighth-grade reading level. Technical, legal or academic copy naturally scores lower.

Is passive voice always wrong? No. One or two instances are normal and sometimes correct. The count exists so the choice is deliberate.

How is this different from the GEO Content Scorer? This editor makes your writing easy for people to read. GEO Content Scorer checks how quotable it is for AI answer engines. Write clearly here first, then score it there.

Does it cost credits? No. Nothing here calls the API. See How credits work.

Can I save the analysis? No. It is live-only and rebuilds from your draft every time you type, so there is nothing to save — the draft itself is kept in your browser on this device.