Dark traffic describes visits whose true source is hidden from analytics. The term dark social covers the most common case: links shared in private channels like Slack, WhatsApp, text messages and email, which pass no referrer. These visits mostly land in Direct. For B2B SaaS, where buyers share links in team chat and internal docs, dark traffic can hide a large part of how people actually discover a product.
Why it exists
Analytics depends on the referrer header and campaign parameters. Native apps, messaging tools, PDFs and many email clients open links without a referrer, and people rarely add UTM parameters when they paste a link to a colleague. The visit arrives with nothing to identify where it came from.
AI assistants add a newer layer. Some send a referrer when users click a cited link, and some do not, and behavior varies by app and platform. A share of AI-driven visits likely sits in Direct too, but the size is not something analytics can measure precisely.
Ways to estimate it
You cannot measure dark traffic directly, but you can find its traces.
- Direct sessions landing on long, deep URLs that nobody types
- Self-reported attribution: a 'how did you hear about us' field on sign-up and demo forms
- Branded search growth in Search Console after a community mention
- Sales call notes that mention a podcast, Slack group or newsletter
What to do about it
Accept that some influence will never be tracked, and add qualitative data to fill gaps. Make sharing easy with clean, short URLs and share buttons that can add campaign tags. Report dark traffic estimates as ranges with the method stated.
Example
A Slack community that never showed up in referrals
Imagine Orchidvane, a developer tooling company, adds a free-text 'how did you hear about us' field to its trial form. A specific engineering Slack community appears again and again, though it never shows in GA4 referrals. Direct sessions to a particular guide rose in the same weeks. The team starts contributing to the community and tracks self-reported mentions as its measure.
Common mistakes
Assigning all direct to dark social
Some direct traffic is genuine typed visits or tracking errors. Treat dark traffic as an estimate, not a reclassification.
Relying on one method
Combine landing page patterns, self-reported data and branded search trends rather than trusting any single signal.
Questions
Dark social is the main cause of dark traffic: private sharing. Dark traffic is broader and includes any visit whose source is lost, such as links in documents or apps.
