How Rankwize measures your AI visibility.
What data we capture, where it comes from, and why our methodology produces results you can trust.
Direct from the source.
Every AI citation tracked by Rankwize comes from a live API call to the actual AI platform — OpenAI for ChatGPT, Google for AI Mode and Overviews, Perplexity, Microsoft for Copilot, and Anthropic for Claude. We don't scrape rendered web pages, simulate model responses, or rely on synthetic data. The same response your customer would see is the response we track.
Daily tracking, every paid tier above $99.
Every prompt you monitor is re-checked on every enabled AI platform daily. We capture the full response text and every citation URL — not just mention counts. Your dashboard shows day-over-day changes, citation flips, and competitor appearances at the same daily granularity.
Tracking runs
Every prompt, every enabled platform, every day.
Diagnosis runs
Only when citation status changed since last cycle.
Safety-net re-diagnosis
Every prompt gets a fresh diagnosis at least every 4 weeks regardless.
Smart re-diagnosis — work happens when things actually change.
Tracking runs daily. Diagnosis runs only when something actually changed in your citation surface — a citation flipped, a competitor appeared, your AI Overview position moved. If nothing changed since last week, we don't waste your AI Credits re-running the matcher and recommendation engine on data that hasn't moved. As a safety net, every prompt gets a fresh diagnosis at least every 4 weeks regardless.
Most AI visibility tools run the full analysis pipeline on every cycle, billing you for work the data didn't need. Rankwize doesn't.
Correlated with your real organic traffic.
Native integrations with Google Search Console (17 endpoints) and Google Analytics 4 (11 endpoints) — uncontested depth in the category — let Rankwize correlate AI citation gains with actual organic clicks and conversions. You see what's moving in AI search alongside what's moving in your traffic and revenue, in one workspace.
What we don't do.
We don't scrape ChatGPT.com, claude.ai, or perplexity.ai.
We don't simulate AI responses with our own models.
We don't cap your tracked responses behind hidden quotas.
We don't run cached or sampled data and call it daily monitoring.
Every data point on your dashboard came from a live API call within the last 24 hours.
Frequently asked questions
How does Prompt Discovery work? ↓
Rankwize fingerprints your domain, indexes your content library, generates candidate prompt intents across buyer-intent stages, scores each on four independent factors (real search demand, citation opportunity, content readiness, commercial value), deduplicates, and routes the top-scoring intents into active monitoring.
What are the four factor scores? ↓
Real search demand (DataForSEO-grounded), citation opportunity (how often AI engines cite anyone for this prompt), content readiness (whether you have a matching page), and commercial value (where the prompt sits in the buyer journey).
What classifies a "citation" vs a "mention"? ↓
Cited: the AI response explicitly named your URL as a source. Mentioned: the response named your brand without citing your URL. Invisible: no reference at all. We track all three separately.
What are the seven fix types? ↓
Update Existing, Add Evidence, Create Comparison, Create How-To, Create Diagnostic, Improve CTA, and Offsite Seed. Each gap gets exactly one primary classification, with up to three secondary recommendations per failing prompt.
What is Gap Escalation? ↓
When a high-priority prompt (priority score 70 or above, or flagged as P1 commercial value) silently degrades from cited to invisible, the system escalates it into a new diagnosis pass instead of letting it sit in the monitoring queue. The escalation carries a causal signal — STALE_CONTENT_RISK, COMPETITOR_PRESSURE, or TRENDING_DOWN.
What is Decay Prediction? ↓
Three causal signals — trending-down citation rate, stale-content risk, and competitor pressure — combine into a per-prompt risk score with a "runs to zero" projection. Lets you intervene before a high-value prompt collapses.
See the methodology in action.
Run a free AI Visibility Report on your domain — live API data, no simulation.