Page views, unique visitors, referrers, and device breakdown (last 30 days). Human traffic only — AI agents are reported separately below.
| Page | Views |
|---|
| Source | Visits | Landing Pages |
|---|
Keywords from search engine referrers. For Google, use Search Console.
| Search Term | Visits | Landing Page |
|---|
Visits from links with ?utm_source=... parameters.
| Source | Medium | Campaign | Visits | Landing Page |
|---|
An AI agent requested the page. Live fetch means a person asked an assistant something and it read your page to answer — that is the citation path. Verified means the source IP matched the operator's published ranges.
Sessions groups an agent's fetches into runs of activity separated by 30 minutes or more — closer to "how many people" than the raw request count, since one question can pull a dozen pages. Shown only for live fetches; for a crawler a run is a crawl, not a person. Requests that failed IP verification are excluded from every count and shown separately, so impersonators do not inflate the totals.
| Agent | Type | Sessions (7d) | 7d | 30d | Code-hub (30d) | Verification (7d) | Top pages |
|---|
A person clicked through to your site from an AI assistant. This is a
floor, not a total — some assistants send no referrer
and no utm_source,
and those arrivals are indistinguishable from direct traffic.
| Source | 7d | 30d | Landing pages |
|---|
Expect zeros for GPTBot, ClaudeBot and Google-Extended.
Cloudflare's managed robots.txt blocks those three training crawlers at the edge — a deliberate
trade documented in docs/austin-code-hub/ARCHITECTURE.md:
training exclusion accepted, live citation preserved. The live-fetch agents above are the ones that matter.
| Page | State | Live for | Last crawl |
|---|
| Query | Clicks | Impr | CTR | Pos |
|---|
| Page | Clicks | Impr | CTR | Pos |
|---|
Search Console data ends about two days back, so these figures do not line up with the live traffic numbers elsewhere on this page.
Which listings get the most views and inquiries (last 30 days)