Channels

The Channels report groups all of your traffic into high-level categories that reflect how visitors found you. Where the Referrers report shows you individual domains, Channels gives you the strategic view — what proportion of your traffic is coming through search, social, paid, or emerging sources like AI assistants.

How channels are determined

Statalog classifies each session into a channel using a two-step process:

  1. UTM parameters — If the landing URL contains UTM parameters, the utm_medium value takes precedence. utm_medium=email maps to Email, utm_medium=cpc or utm_medium=paid maps to Paid Search, and so on.
  2. Referrer domain categorisation — If no UTM parameters are present, Statalog matches the referrer domain against built-in lists of search engines, social networks, AI assistants, and other known categories.

This means channels work automatically with zero configuration for most traffic. UTM tagging is only needed when you want to override the automatic classification — most commonly for email campaigns and paid advertising.

The channel categories

Organic Search Visits from search engines where no paid UTM tag is present. Includes Google, Bing, DuckDuckGo, Yahoo, Ecosia, Brave Search, and others. This channel reflects your SEO performance.

Direct Visits with no referrer information — typed URLs, bookmarks, native app links, and cases where the referrer was stripped by the browser or client. See the Referrers report documentation for a full explanation of why direct traffic is often larger than expected.

Referral Visits from external websites that are not search engines, social networks, email providers, or AI assistants. These are links from other websites — press coverage, directories, partner sites, community posts.

Social Visits from known social networks: Twitter/X, Facebook, Instagram, LinkedIn, YouTube, Pinterest, TikTok, Reddit, and others. Statalog's social domain list is maintained and updated as new platforms emerge.

Email Visits tagged with utm_medium=email, or from known email service provider referrer domains. If you send newsletters without UTM parameters, those clicks will appear as Direct. Adding utm_medium=email to your newsletter link URLs correctly routes them into this channel.

Paid Search Visits tagged with utm_medium=cpc, utm_medium=ppc, utm_medium=paid, or similar paid search conventions. This channel represents your paid advertising spend on search platforms.

AI Assistants Visits originating from AI chat tools including ChatGPT, Perplexity, Claude, Microsoft Copilot, Gemini, and other AI-powered interfaces. Statalog detects these by matching referrer domains from known AI assistant platforms.

Why AI Assistants is a channel most tools miss

As AI-powered chat interfaces become a significant source of web traffic, most analytics platforms have not yet adapted their channel taxonomies to account for them. Traffic from ChatGPT or Perplexity would typically fall into "Referral" in most tools — a catch-all that buries the signal.

Statalog surfaces AI assistant traffic as a distinct channel so you can see it clearly and watch it trend over time. For content-heavy sites, documentation, and product pages, AI-driven referrals are growing rapidly and understanding their share of your traffic is increasingly important.

Comparison with GA4

Google Analytics 4 has a channel grouping system, but it requires significant configuration to work correctly. GA4's default channel groups depend on UTM tagging conventions that many teams do not follow consistently, and the "AI Assistants" channel is absent from GA4's default taxonomy.

Statalog auto-classifies channels without requiring any configuration. The classifications are applied retroactively to all historical data, so adding a new traffic source does not leave gaps in your channel history.

FAQ

What if a session does not match any channel? Sessions that do not match any known pattern are grouped into an "Other" channel. This is typically a small residual category. If you see significant "Other" volume, check whether a traffic source is using unusual UTM values that do not match the standard conventions.

Can I define custom channels? Custom channel definitions are not currently available. If you have a specific channel mapping need, UTM tagging is the most reliable way to control how a traffic source is classified — use utm_medium values that match standard conventions (email, cpc, etc.) and Statalog will classify them correctly.