Live visitor intelligence
Visitor monitoring gives operators the context they need to respond intelligently: where someone came from, what they are viewing, how long they have stayed, and whether they have chatted with you before. Because ZChat is self-hosted, that behavioral data stays under your control.
Review pages viewed, referral source, region, browser, and repeat-visit history before the first reply is sent.
Use visitor behavior to decide when outreach makes sense instead of blasting everyone with the same invitation.
Support sees troubleshooting context while sales sees page-level intent and product interest in real time.
Track current visitors, available operators, waiting conversations, and recent sessions from the same admin workspace.
Most support tools only show you someone once they have already asked for help. Visitor monitoring shows the people who are still deciding — and gives your team enough context to be useful rather than generic.
The page a visitor is currently viewing, so an agent answering a question about billing already knows the person is sitting on the pricing page rather than the documentation.
The referring site, so a visitor who came from a comparison article can be treated differently from one who typed your address directly.
A visit count, so a third-time returner reads as a warmer prospect than a first-time arrival who may just be browsing.
Device and browser, which turns "it doesn't work" into a question you can actually answer, plus approximate location from the connection.
Browsing and chatting share one durable visitor identity, so a person who has been reading your docs for ten minutes and then starts a chat appears to the agent as one visitor with history — not as an anonymous browser and a separate unrelated chat.
Seeing a visitor is only useful if you can act on it. From the visitor list an operator can send a personal invitation with a message written for that moment, and the visitor gets a banner they can accept or decline. Accepting carries the agent's opening line into the chat, so the conversation starts mid-thought instead of at a blank form.
A visitor who is already invited, waiting, or in a conversation cannot be invited again, so nobody is greeted twice. If you would rather this happened automatically, proactive chat triggers cover the same ground without an operator watching.
Watching people browse is a capability that deserves a deliberate decision, so ZChat does not make it for you.
No. It is off until an administrator enables it in settings. Until then the widget does not report browsing activity at all, and the visitor list only shows people who have started a chat.
Yes. Add one attribute to the embed on that page and it is excluded, even while monitoring is enabled everywhere else. This is the usual approach for checkout, account and other sensitive pages.
In your own SQL Server database, on your own server. ZChat does not send visitor activity to any external service, and the records are covered by the same data export and erasure tools as your chat transcripts.
The widget sends a small periodic beacon at an interval you control, and the interval is clamped to a sensible range. It is not a continuous stream, and it stops when the visitor leaves the page.
ZChat gives you the installable server, web dashboard, website widget, and desktop agent tools in one self-hosted product you buy once and keep. Run it on infrastructure you trust and connect AI only if and how you want it.
Deployment
Install on Windows or Linux, behind IIS or Nginx, in a VM, or in Docker if that fits your stack.
Commercial model
One-time purchase, perpetual license, and no monthly per-agent bill attached to growth.
AI Flexibility
Use Ollama locally or connect OpenAI and Anthropic with your own provider accounts.