Direct answer
FreeScan converts public evidence into rule points, category scores, and an overall score.
FreeScan calculates each category score by dividing earned rule points by available rule points, then converting the result to a 0 to 100 scale. A passed check earns all available points, a failed check earns none, and an unknown check earns 40% because the safe public scan could not verify the signal.
The overall score is the rounded arithmetic average of four category scores: SEO / AEO, security, accessibility, and design. Each category has equal weight in the overall score. Checks inside a category can carry different rule-point values.
The formula
Every score starts with the status and point value of each check.
A check asks a defined question about the scanned page or a related public resource. Examples include whether a canonical URL is present, whether the page uses HTTPS, whether an image has usable alternative text, or whether the rendered page exposes serious accessibility violations. The check has a maximum rule-point value based on its importance within that category.
| Check status | Points earned | Meaning |
|---|---|---|
| Pass | 100% of the check's rule points | The public evidence satisfied the check. |
| Fail | 0% of the check's rule points | The public evidence did not satisfy the check. |
| Unknown | 40% of the check's rule points | The safe public scan could not verify the signal. |
Category score = round(earned rule points ÷ available rule points × 100)
Overall score = round((SEO + security + accessibility + design) ÷ 4)
Why unknown earns partial credit
Missing evidence should not be reported as a pass or a failure.
Some signals cannot be verified during every safe public scan. A public file may be unreachable, a rendered browser audit may not complete, or a response may not provide enough evidence for a confident result. FreeScan labels that check unknown and explains what could not be confirmed.
Unknown earns 40% of the available points. This prevents an unavailable signal from receiving full pass credit, but it also avoids treating missing evidence as a proven defect. An unknown check still appears in the report because it may require a manual review or a later rescan.
The four categories
Each category answers a different website health question.
A single combined score is useful for comparison, but it can hide the reason a page needs work. The four category scores keep separate technical and user-facing concerns visible.
| Category | What FreeScan reviews | Why it matters |
|---|---|---|
| SEO / AEO | Crawl access, metadata, headings, content signals, structured data, public discovery files | Search engines and answer systems need clear, reachable, consistent information before they can understand or cite a page. |
| Security | HTTPS, public headers, forms, exposed files, mixed content, browser-visible risks | Weak public security signals can create avoidable risk, warning states, broken trust, and unsafe data handling. |
| Accessibility | Semantics, labels, headings, images, keyboard support, contrast, rendered violations | Accessibility defects can prevent people from reading, navigating, understanding, or completing an action. |
| Design | Hierarchy, calls to action, readability, responsive layout, runtime health, performance signals | A page can be technically available while still making the next step hard to find, understand, or complete. |
SEO / AEO
This score measures whether public systems can reach, interpret, and represent the page.
Search visibility begins with access and interpretation. FreeScan reviews public signals such as crawler access, canonical URLs, titles, descriptions, headings, structured data, sitemaps, and answer-ready content. These checks help identify technical barriers and missing context that can reduce a page's ability to be indexed, understood, or presented accurately.
A higher score does not guarantee a ranking. Query intent, competition, content quality, links, site reputation, location, freshness, and many other systems still affect search results. The score is useful because it identifies page-level work that is under the site owner's control.
Security
This score reviews public protection and browser trust signals without attempting an exploit.
FreeScan checks signals that can be evaluated safely from the public web. These can include HTTPS, common security headers, mixed content, form behavior, and exposed public files. A failure can identify a configuration that deserves review before visitors submit information or interact with the page.
The scan does not log in, test private routes, attack the application, or replace a professional penetration test. A high score means the scanned public signals were strong. It does not prove that the application, infrastructure, dependencies, or internal controls are secure.
Accessibility
This score finds public barriers that automated and rendered checks can detect.
FreeScan reviews page structure and rendered behavior when browser evidence is available. Checks can cover labels, headings, landmarks, images, contrast, keyboard-relevant controls, and violations reported by automated accessibility testing. These issues matter because they can block people from understanding content or completing a task.
Automated testing finds only part of the accessibility work. Manual keyboard review, screen-reader testing, content review, zoom and reflow testing, and evaluation against the applicable WCAG success criteria are still necessary. The category score is a useful defect signal, not a certification.
Design
This score measures clarity, usability, responsive behavior, and public-page quality signals.
Design checks focus on whether a visitor can understand the page and find the intended next step. FreeScan reviews defined signals such as heading hierarchy, call-to-action clarity, readability, mobile layout, rendered errors, visual stability, and performance data when it is available.
These signals can affect comprehension and conversion because unclear copy, broken layouts, runtime errors, and slow responses create friction. The score does not judge visual taste, brand originality, customer research, or the complete conversion path after the scanned page.
Recommendation order
The biggest score loss is important, but it is not the only priority signal.
FreeScan ranks failed and unknown checks that have remediation guidance. The ranking combines four inputs so the first recommendation is not selected by raw point loss alone.
Severity
Critical and high-severity findings receive more urgency than medium, low, or informational findings.
Normalized score loss
Points lost are compared with all available points in that category, which keeps differently sized categories comparable.
Category urgency
SEO and security receive a 1.15 urgency weight. Accessibility receives 1.05. Design remains at 1.0.
Remediation priority
Every remediation has a priority number. Lower numbers are more urgent and add more weight to the recommendation order.
This order is a starting point. A team may need to change it when a finding affects a regulated workflow, active incident, major revenue path, release dependency, or known user barrier.
Fix prompts
A fix prompt turns one finding into a scoped implementation brief.
Each eligible failed or unknown check includes a copyable fix prompt. It carries the report evidence into a structured task so a developer or coding agent can investigate the correct issue without receiving the entire report as unstructured context.
| Prompt section | What it contains |
|---|---|
| Context | The public URL, domain, category, check ID, status, severity, priority, and score impact. |
| Evidence | What FreeScan observed, what needs attention, and why the issue matters. |
| Implementation scope | Likely files or systems to inspect, the requested change, and check-specific guidance. |
| Acceptance criteria | The public result that must improve and the regressions that must not be introduced. |
| Verification | A requirement to reproduce the issue, make the smallest appropriate change, test it, deploy it, and rescan. |
| Safety | Instructions to protect credentials, private data, security controls, accessibility, existing routes, and rollback options. |
A fix prompt does not change the website automatically. The recipient still needs the repository, CMS, server, CDN, or hosting context that controls the public result. The prompt explicitly requires verification before editing, a narrowly scoped change, regression checks, and a rescan after deployment.
Score changes
A new scan measures the public page and the current check library again.
A report records what FreeScan could verify from the public URL during that scan. A score can change after a deployment, content edit, header change, hosting change, browser-rendered result, or update to a related public file. It can also change when FreeScan adds a check or adjusts rule points to reflect the current methodology.
Compare the scan version, category totals, check evidence, and completed time when reviewing reports from different dates. The useful question is not only whether the total moved. Check which public signals changed and whether the change improved the real visitor or crawler outcome.
Limits
Use the score as a diagnostic summary, not as proof of an outcome.
- The score covers one public URL, not every page, template, user state, or authenticated workflow on a domain.
- The SEO score is not a Google ranking factor and cannot predict a position for a query.
- The security score is not a penetration test and does not attempt exploits or authenticated access.
- The accessibility score is not WCAG certification and cannot replace manual assistive-technology testing.
- The design score measures defined public-page signals, not brand taste or complete conversion research.
- A score of 100 means all current rule points were earned for the scanned page. It does not mean the website is perfect.
Scores are most useful when compared with the evidence beneath them. A 72 caused by one severe security failure requires a different response from a 72 caused by several lower-impact design issues. The category breakdown and check details provide that context.
Review workflow
Use the report to decide what to verify, fix, and rescan.
Step 1
Read the four category scores
Compare SEO / AEO, security, accessibility, and design before using the overall score as a summary.
Step 2
Open the checks that lost points
Review the status, evidence, severity, rule points, and explanation for each failed or unknown check.
Step 3
Start with ranked recommendations
Use the recommendation order to focus on severe issues with meaningful score impact and an urgent remediation priority.
Step 4
Use the fix prompt with implementation context
Copy the fix prompt into a coding agent or hand it to a developer together with the repository, CMS, hosting, and deployment context needed to verify the issue.
Step 5
Deploy and rescan the public URL
Verify the real user outcome, deploy the change, and run another scan to confirm that the public signal improved without creating regressions.
FAQ
FreeScan website audit scoring FAQ
How is the FreeScan website audit score calculated?
Each check has available rule points. Passed checks earn all points, failed checks earn none, and unknown checks earn 40%. FreeScan divides earned points by available points within each category and converts the result to a 0 to 100 score. The overall score is the rounded average of the four category scores.
Are the four FreeScan category scores equally weighted?
Yes. The overall score gives equal weight to SEO / AEO, security, accessibility, and design. Checks within a category can have different rule-point values, so a higher-value check affects its category score more than a lower-value check.
What does an unknown check mean?
Unknown means the safe public scan could not verify a signal with enough confidence. It does not mean pass or fail. FreeScan keeps the result visible, awards 40% of the available points, and explains what could not be confirmed instead of inventing evidence.
Is a FreeScan score a Google ranking factor?
No. Google does not use a FreeScan score. The SEO / AEO score summarizes crawl, metadata, content, structured data, and public discovery signals that can affect search visibility when they are implemented correctly.
Does a score of 100 mean a website is perfect?
No. A score of 100 means the scanned public page earned all currently available rule points. It does not prove perfect rankings, complete security, full accessibility, strong conversion, or quality across every page on the domain.
Why can a website audit score change?
A score can change when page content, metadata, headers, rendered behavior, public files, browser output, or the FreeScan check library changes. A new scan measures the current public response, so cached or undeployed changes may not appear yet.
How does FreeScan rank website audit recommendations?
FreeScan combines severity, the percentage of category points lost, category urgency, and remediation priority. SEO, security, and accessibility receive modest urgency adjustments. Lower remediation priority numbers indicate more urgent work.
What is included in a FreeScan fix prompt?
A fix prompt includes the scanned URL, check ID, category, status, severity, priority, score impact, evidence, why the issue matters, likely implementation locations, a scoped task, acceptance criteria, verification requirements, and safety rules.
Does a fix prompt change the website automatically?
No. A fix prompt is a structured task brief. A developer or coding agent still needs access to the correct repository, CMS, server, CDN, or hosting configuration, and must verify the issue before making and deploying a change.
Is the security score a penetration test?
No. FreeScan reviews public security signals that can be checked safely, such as transport, headers, exposed files, forms, and browser-visible behavior. It does not perform authenticated testing, exploit attempts, or a full penetration test.
Is the accessibility score a WCAG certification?
No. FreeScan combines automated and public-page accessibility checks, including rendered signals when available. Manual testing with keyboard users, screen readers, content review, and the relevant WCAG success criteria is still required.
Source notes
Public standards explain the signals. FreeScan defines the scoring model.
FreeScan's point values, status treatment, category formula, overall formula, and recommendation ranking are documented on this page. The following primary references provide additional context for the public signals reviewed by the audit.
Summary
FreeScan scoring methodology in seven statements
- A FreeScan category score is earned rule points divided by available rule points, expressed from 0 to 100.
- Pass earns all points, fail earns no points, and unknown earns 40% because the signal could not be verified safely.
- The overall score is the rounded average of four equally weighted category scores.
- Rule-point values differ by check, so a higher-value check has more influence inside its category.
- Recommendations use severity, normalized category score loss, category urgency, and remediation priority.
- Fix prompts turn one audit finding into a scoped implementation and verification brief.
- A FreeScan score summarizes public evidence. It is not a ranking factor, certification, penetration test, or outcome guarantee.
