Web analytics

First-party visitors and revenue for one app, plus the older listing report that still reads a linked GA4 property.

First-party (Web analytics)

Open Web analytics in the sidebar. The page is visitors beside revenue: conversion, revenue per visitor, bounce rate, average session time, and who is on the site now. Click a breakdown row to keep only that source, page, or place. The arrows step the current window backward and forward. Pin a range when you want it back later.

Revenue is the payments you post (Stripe Checkout handoff, the payments API, or a payment event), minus refunds. A customer who never paid is not treated as churned. Billing charges that never come through one of those paths stay on the revenue pages; this page does not invent a match for them.

Install the snippet

On Setup, copy the stack you use:

  • Browser SDK — paste /sdk.js on every page. Page views, route changes, engaged time, outbound clicks, and data-corex-goal goals are recorded by the same snippet that already sends usage events. There is no second tracker.
  • Shopify theme app embed — the same script in the theme embed, turned on in the theme editor.
  • Google Tag Manager — a custom HTML tag on all pages.
  • Next.js — load /cx/sdk.js and rewrite /cx to your CoreX host.
  • Cloudflare proxy — serve /cx/* from your own hostname so the script is first-party.
  • Stripe Checkout handoff — put visitor_id and session_id on the Checkout metadata (or call CoreX.checkoutUrl before redirecting) and add ?payment_id= on the success URL. That return only attaches the visitor to a payment you already posted. It does not set the amount. A Shopify install link gets the same cx_vid, cx_sid, and ref query.
  • Copy AI install prompt — paste it to an assistant working in the project.

Add ?corex_ignore=1 once in your own browser. That sets a cookie so your visits are dropped. You can also list office addresses under Settings.

Server-side goals and payments (Paddle, Lemon Squeezy, or anything else) use the same API token as usage events:

POST /api/goals     { "name": "signup", "visitorId": "…" }
POST /api/payments  { "paymentId": "…", "amountCents": 4900, "refundedCents": 0, "visitorId": "…" }

paymentId is required. Posting the same id again updates that payment and returns 200. customerId is a shop domain, a Shopify shop id, or this app's customer id — the same lookup as a usage event. A browser CoreX.payment() call does not write revenue.

The same paths exist under /api/v1. Name the app when the token is not already scoped to one. The ignored-address list is one IP or one CIDR range per line (203.0.113.9 or 203.0.113.0/24), not a text prefix.

Privacy

Cookieless mode replaces the visitor id with a daily hash of a salt, the date, the address, and the browser. The same person is a new visitor the next day. Counts stay useful; journeys get shorter. The Settings card says this before you turn it on.

Set a cookie domain (.example.com) to follow a visitor across subdomains. List every hostname that should count, including a sister domain, under allowed hostnames. Leave that list empty to accept every site that has the snippet.

Share, import, alerts

A share link is a read-only page at /w/{token}. The token is shown once and is not indexed. Import a Google Analytics or Plausible spreadsheet (a date column and a visitors column) to fill days from before the snippet. Traffic drops and source spikes are rules on the existing revenue alerts, not a second alert product.

The customer page shows the same visit as a journey and links to the replay. Funnels stay in listing analytics; first-party event names that have actually arrived are added to that builder.

Listing analytics (GA4)

What you see

Listing Analytics (per app) reports sessions, visitors, traffic sources and listing conversion from the GA4 property you link to that app, plus a live "on the listing right now" count and a pageview → install funnel.

Link the property once under the app's settings; CoreX caches each report per app and range, and Refresh forces a re-pull when you need it now.

Use it

  • Compare source-level sessions against installs to find the channel that converts rather than the one with the most traffic.
  • Read it beside Keywords: rank moves show up in sessions before they show up in installs.

Why it matters

One acquisition view — web traffic and App Store outcome — beats flipping between GA and CoreX and eyeballing the join.