Open methodology
We grade websites on traditional SEO and AI-search readiness using an open algorithm. Every check has a fixed ID, every category has a published weight, and every AI-generated recommendation is grounded in those checks. No black box, no made-up metrics.
GEO (Generative Engine Optimization) is the discipline of making content discoverable, parseable, and citable by AI search systems — ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini. Traditional SEO ranks you in a list of links; GEO gets you quoted in the answer.
The overall score is a weighted average of category scores. Weights are loaded from the same code path that runs the audit, so this table never drifts from production.
| Category | Weight | Checks |
|---|---|---|
| Content | 23% | 15 |
| Technical SEO | 22% | 26 |
| On-Page SEO | 18% | 25 |
| Structured Data | 14% | 22 |
| Performance | 12% | 11 |
| AI Search (GEO) | 8% | 16 |
| Images | 3% | 13 |
| Total | 100% | 169 |
Some categories run checks but do not contribute to the headline score yet: Security, Accessibility, Frontend.
Every AI-generated bullet — in the summary, in the developer brief, in the action plan — must reference at least one real check ID with its status, e.g. tech.canonical (FAIL) or schema.faq (MISSING). The model is instructed to quote the actual current title, current meta description, and actual headings of the audited page, not synthesize generic advice.
After the model responds, we validate the output for grounded references. If a summary fails to cite any real findings, we retry once with stricter instructions. We would rather ship a slightly less polished sentence than a confident hallucination.
Every check is a pure function of the data we collect — given the same crawl, it always returns the same pass/fail/warn. What changes between runs is the input: your site, the PageSpeed Insights measurement, and the availability of external sources we query during the crawl.
We measured it. Across 174 repeat audits of the same domains on the same day, the overall score varied by 1.5 points on average and the AI-search (GEO) score by 3.6 — enough to move a borderline grade by one letter, not enough to change the picture. Performance metrics (LCP, CLS, INP) come from Google PageSpeed Insights and reflect real-time measurements. AI-generated text (summary, developer brief, 30-day plan) runs at temperature 0: the same facts, check IDs and recommendations appear every time, but DeepSeek is not byte-deterministic, so phrasing can differ by a few words.
Every check that runs against your site, grouped by category. Severity reflects the engine's default — individual results may upgrade or downgrade based on what's found.
Spotted a check that feels wrong? Email info@seoport.com.ua