Observed evidence
A dated observation from a configured provider, submitted answer snapshot, crawl response, page fetch, schema check, sitemap entry, or llms.txt file. The report should preserve the source and limitation beside the conclusion.
Methodology
How VisAI keeps AI visibility diagnostics evidence-safe.
VisAI separates observed evidence, crawl readiness, citation readiness, and unmeasured sources so a SaaS team can act on GEO gaps without turning a limited scan into a ranking claim.
Crawl readiness
Whether public pages can be discovered, fetched, rendered, and parsed. This includes robots.txt, status codes, sitemap coverage, server-rendered product copy, and basic access blockers.
Citation readiness
Whether public pages contain sourceable product facts, buyer questions, proof, comparison detail, docs, FAQ answers, and trust signals that a diagnostic can point back to.
Not measured
A deliberate label when a source, prompt, provider, citation, or answer platform was not captured with usable evidence. Not measured is safer than pretending absence or presence was proven.
Report states
What each state means before someone pays
Starter Snapshot checkout stays tied to report readiness. If evidence is not strong enough, the product should explain what to fix or re-scan instead of pushing a payment.
Free Snapshot Preview
Fast triageShows crawl blockers, core readiness signals, first fixes, and visible evidence limits for a public domain. It is useful for deciding what to fix or whether stronger evidence is worth collecting.
Limited Diagnostic
Useful but scopedThe scan has some actionable evidence, but answer-source coverage, citations, category confidence, or public-page quality is not strong enough for a full delivery claim.
$99 Starter Snapshot
Unlocked when evidence is readyA protected founder-ready handoff with prompt snapshots when evidence exists, competitor context from usable observations, evidence table, top fixes, and downloadable delivery.
Claim boundaries
What a VisAI report can and cannot say
Say
This scan found crawler blockers on public pages.
Say
Provider evidence may be limited, and unmeasured sources are labeled.
Say
The next fix is to publish a crawlable comparison page and re-scan.
Do not say
This guarantees a higher ChatGPT, Perplexity, Claude, Gemini, or Google AI Overview position.
Do not say
A readiness score proves platform-wide AI visibility.
Do not say
Submitted-only evidence or a submitted-only answer snapshot makes the report Ready by default.
Three questions Starter expands
When a report has enough evidence for protected delivery, Starter Snapshot turns the free diagnostic into a clearer founder or agency handoff.
- 1Which configured prompts produced usable answer or citation evidence, and which remained unavailable?
- 2Which competitors appeared in usable observations, and what public pages could explain the difference?
- 3Which top fixes should founder, SEO, content, and dev owners ship first, and how should each fix be validated?
Protected delivery never changes the evidence boundary. It adds handoff depth only when the report is ready enough to support it.