Why might a repeat score fall when I did not change the site?
First establish the number’s scope and available evidence. Then distinguish a confirmed failure, manual review and missing measurement; each state needs a different next step.
When comparing scans, open the PDF reports and compare rule identifiers. A robots.txt change can affect the evidence available for technical SEO; an unavailable speed measurement belongs in a different queue from a confirmed canonical URL defect.
What to do step by step
- Levels one to three represent confirmed problem severity.
- Level four is manual review without an automatic deduction.
- Level five represents a passing rule.
- Repeated occurrences do not become separate rule deductions.
- SEO and AI views use checks in their respective scopes.
A practical example
Failures receive 2.25 times the weighting of passes and a minimum deduction of 4.5 points per weight unit. A confirmed critical rule has a 40-point minimum before total loss is capped at 100 and apportioned when necessary. The audit shows specific deductions.
A common mistake
A score is not simply the percentage of checks passed, nor the probability of ranking first. Different tools may use different definitions.
Try the report before checking your site.
Explore the sample report, then enter your own URL to see findings, HTML evidence and a repair plan for your page.
Run your audit
What do the check statuses mean?
Failed means the available rule’s condition was violated. Review needs human context. Unavailable evidence and an inapplicable check have separate states.
- Do not interpret a missing performance measurement as a fast page.
- Do not treat a check for an unnecessary element as a requirement to add that element.
How does a score differ from a task count?
Passed and failed checks with positive weights enter the rule score. Each rule identifier counts once; occurrence counts help establish the repair scope.
- A hundred occurrences of one defect may need one template edit followed by broad regression checks.
- Advice with zero weight can still be useful even though it does not change the number.
Which scores can you compare?
The report separates technical areas, content signals, the query and performance. The Fukamo score follows its own rules; it is not a rating assigned by Google.
- Keep the URL, query, measurement type and available evidence consistent when comparing results.
- Explain a changed total through specific rules, rather than the point difference alone.
How to prioritise SEO repairs
More severe failures have a larger impact in the methodology. Add page purpose, affected templates and dependencies when deciding the work order.
- Resolve accidental blocking of a public service page before polishing its wording.
- After fixing a failure, verify fresh evidence and usability; a high score alone does not establish orders or indexing.
How to verify the result
Use evidence, impact and the recommended fix to decide. Retest rule IDs on the same URL; measure actual search performance and AI citations separately.
Primary documentation developers.google.com ↗
Frequently asked questions
Why might a repeat score fall when I did not change the site?
Compare available inputs, server responses, the exact URL, and rule versions. Dynamic content or rejected requests may change what can be verified. Interpret the score alongside findings; a small point difference does not establish a ranking decline or a worse business.
When the task is complete
Comparisons retain the same scope and explain new or unavailable evidence. Fix decisions follow severity and cause rather than a desire to reach a round number.
Related SEO guides
Check a specific page.
Fukamo shows findings, evidence and the next step.
679 rules + SEO, AI readability, performance and accessibility measurements for free.