SEOBlitz

Published by DigitalRichKid. Maintained against shipped v2.0 behavior. Report a correction.

What the model evaluates

SEO Blitz reviews a supplied page title and body draft. The selected content mode changes the minimum depth, required lexical cues, some structure expectations, and the weight of each dimension. The result contains eight dimension scores, one rounded overall score, findings, and three prioritized revisions.

The eight dimensions are Clarity, Completeness, Structure, Readability, Specificity, Repetition, Title alignment, and Publication readiness.

How the score is calculated

  1. Each dimension begins at 100.
  2. Every triggered rule subtracts its configured deduction from its dimension.
  3. Each dimension is clamped at zero. Deductions can still produce findings after the dimension reaches zero, but they cannot reduce that dimension further.
  4. The selected mode applies one weight to each of the eight clamped dimension scores.
  5. The weighted values are added and the final score is rounded once to the nearest integer, then clamped from 0 to 100.
dimension = max(0, 100 - sum(deductions))
overall = clamp(0, 100, round(sum(dimension * mode weight)))

All six current weight sets total 1.00. Findings are ordered by severity, then deduction, then the active mode weight for the finding's dimension. High severity means 12 or more points, Medium means 6 to 11, and Low means 0 to 5. Severity describes a configured deduction, not measured SEO harm.

Six Content Modes

ModeMinimum wordsStructure expectationRequired cue groups
Blog article or guide300At least 3 paragraphs and, above 150 words, at least 2 Markdown-like headingsReader problem, example, proof or detail, next step
Landing page200At least 2 paragraphsOffer, deliverable, fit, call to action
Product description120At least 2 paragraphsSpecifications, use case, caveat
Local service page200At least 3 paragraphsService area, problem, process, contact, scope
Help article150At least 2 paragraphs and ordered-step line cues above 80 wordsSteps, prerequisites, outcome
Comparison page200At least 3 paragraphs and, above 150 words, at least 2 Markdown-like headingsOptions, criteria, fairness, recommendation

Mode cues are case-insensitive word and phrase patterns. A cue match does not prove that the draft covers the subject well. An equivalent unconfigured phrase can also produce a false finding.

Eight Dimensions

DimensionImplemented checksImportant boundary
ClarityConfigured hedging phrases and up to three sentences over 25 wordsNo passive-voice or jargon detector
CompletenessUnder 50 words, under the mode minimum, and missing mode cue groupsCues are lexical, not semantic proof
StructureMarkdown-like heading lines, paragraph count, Help step lines, and paragraphs over 250 charactersNo rendered HTML hierarchy or visual-layout inspection
ReadabilityAverage sentence length, long-sentence count, very-short-sentence majority, and average characters per wordNo Flesch, Flesch-Kincaid, syllable, grammar, or comprehension score
SpecificityConfigured generic claims, digit-plus-unit detail matches, and reader-question term overlapNo factual, citation, example-quality, or semantic verification
RepetitionDominant eligible term and repeated exact three-word sequencesNo conceptual similarity or ideal keyword-density calculation
Title alignmentSupplied title presence, length, body overlap, repeated title terms, and reader-question overlapNo live title tag or title-to-H1 comparison
Publication readinessTitle, mode depth, risky guarantee phrases, limitation cues, and link or next-step cuesNo metadata, author, date, live-link, or page inspection

Overlapping deductions

One draft characteristic can trigger more than one rule because rules can protect different dimensions. Long sentences can affect Clarity and Readability. A short draft can affect Completeness and Publication readiness. A missing title affects Title alignment and Publication readiness. Help mode uses separate completeness and structure expressions for steps. This is intentional v2.0 behavior. It is not independent evidence of ranking impact.

Inputs and display-only metrics

Field or outputRole in v2.0
Content modeScore-affecting. Selects weights, minima, cue checks, and structure expectations.
Main contentScore-affecting and required for a report.
Page titleScore-affecting when present or absent.
Primary reader questionOptional and score-affecting for Specificity and Title alignment when its eligibility conditions apply.
Intended readerPlanning context. Stored and exported, but not scored.
Desired reader actionPlanning context. Stored and exported, but not scored.
Estimated reading timeDisplay-only. Calculated as at least 1 minute, using 225 words per minute. It does not affect the score.

Known boundaries

SEO Blitz analyzes submitted text. It does not crawl a live page, inspect a rendered DOM, validate links, inspect images or alt text, read Search Console or SERP data, verify facts, measure backlinks, or predict rankings. A high score means the draft triggered fewer or lower-weight deductions in this model. It does not prove that the page is accurate, helpful, original, technically sound, or likely to rank.

Detailed documentation

Model Changelog

SEO Blitz Scoring Model v2.0 is the current model. The Version History separates verified repository implementation evidence from unknown public deployment history.

Privacy Model

The current calculator runs in the browser without an account requirement. SEO Blitz stores the current analyzer draft in browser localStorage. In the published Browser Privacy Verification study, submitted test canaries were not observed in captured request URLs, relevant headers, or request bodies during two controlled Chromium runs. Read the Privacy Policy for the broader technical disclosures and scope limits. The study does not establish all hosting, browser, extension, or future-version behavior.

Correction Process

Methodology issues can be reported through Contact. Include the page, issue, expected behavior, actual behavior, and a short non-confidential sample when the issue involves scoring.