Word count measures size, not quality
A 2,000-word article can be rigorous or padded. A 400-word instruction can be complete or unusably thin. Word count helps estimate scope, reading time, print length, and production effort; it cannot judge accuracy, coherence, specificity, or usefulness.
Sentence length reveals rhythm and load
Average sentence length hides distribution. Review median and maximum length, multi-clause sentences, clusters of equally short sentences, and technical sentences that introduce several concepts at once. Short is not automatically clear; long is not automatically difficult.
Paragraph density affects navigation
Online readers use headings and paragraph breaks to locate information. Split a paragraph when its function changes—claim, evidence, example, limitation, or instruction. Avoid both walls of text and a stream of one-sentence fragments.
Readability formulas are proxies
Common formulas use sentence length and word or syllable complexity. They can flag unusual density but cannot judge domain knowledge, logical structure, or whether a technical term is necessary.
Replace avoidable complexity, not required precision. Define necessary specialist language.
Repetition has different meanings
Frequency tools reveal accidental repetition and terminology inconsistency. They do not define an ideal density. Google warns against keyword stuffing.
- Necessary term: standard name of the subject.
- Consistency term: preferred label used throughout.
- Rhetorical repetition: deliberate emphasis.
- Filler: repeated because evidence is missing.
- Search stuffing: variants inserted primarily for ranking.
Use Jivaro’s analyzer
- Open Word Counter & Text Analyzer.
- Paste a complete draft or representative section.
- Review words, characters, sentences, paragraphs, and reading time.
- Inspect readability and sentence-length signals.
- Review repeated terms in context.
- Use cleanup/case tools only on a copy.
- Export a report for comparison or review.
The app cannot decide whether a claim is true or whether the article answers the reader’s task.
Translate signals into edits
| Signal | Question | Action |
|---|---|---|
| Long introduction | How quickly is the query answered? | Move thesis and answer earlier |
| Repeated keyword | Is repetition necessary? | Keep precise terminology; remove stuffing |
| Long paragraph | Does it contain several functions? | Split claim, evidence, example, limitation |
| Dense sentence | Are several judgments combined? | Separate or define terms |
| High word count | Does every section add value? | Remove repetition and unsupported generalities |
Use two editing passes
Use metrics to locate likely friction.
Read for argument, accuracy, evidence, sequence, and audience.
Do not change every flagged sentence automatically.
What metrics cannot tell you
- Whether sources support the claim.
- Whether advice is safe.
- Whether counterarguments are missing.
- Whether examples are representative.
- Whether tone fits the audience.
- Whether search intent is satisfied.
Compare revisions, not isolated scores
Metrics become more useful when comparing a draft before and after editing. Did the maximum sentence length fall while technical meaning remained? Did repeated filler decrease? Did headings make the page easier to scan? Did the final word count shrink because redundancy was removed rather than evidence?
Store a report with the revision or pull request when editorial changes are substantial. The trend is more informative than one absolute score.
Combine analysis with source review
A clear article can still be wrong. Pair the text report with a claim check: identify factual statements, map them to sources, verify dates and quantities, and flag inference. This prevents readability work from polishing unsupported claims into more persuasive misinformation.
Frequently asked questions
There is no universal target. Compare drafts for the same audience and investigate unusual density.
No. Use terms naturally and satisfy the query; artificial repetition can resemble keyword stuffing.
No. Correctness and completeness require source and subject review.
No. First classify whether repetition is necessary terminology, emphasis, inconsistency, filler, or stuffing.
Related Jivaro apps
Count words and characters, estimate adjustable reading and speaking times, review Flesch and ARI readability, inspect sentence balance and repeated keywords, and export reports locally.
Open appSources and references
- Unicode Text SegmentationUnicode Consortium · reference
- Intl.SegmenterMDN Web Docs · reference
- Spam policies for Google web searchGoogle Search Central · reference
- Word Counter & Text AnalyzerJivaro · first-party