What is dark traffic?
Dark traffic is genuine human traffic whose origin analytics cannot see, because the referrer header was stripped and no campaign tag was present. It shows up as direct traffic. The usual sources are private messaging apps, email clients, in-app browsers and privacy settings that suppress the referrer.
Worked example
Estimating dark traffic on a content site.
| Direct sessions in the period | 4,120 |
|---|---|
| Direct sessions landing on the homepage or a short URL | 1,380 |
| Direct sessions landing on a deep article slug | 2,740 |
| Estimate | 2,740 sessions of dark traffic |
About 66% of direct traffic is dark rather than typed Those 2,740 sessions came from somewhere real: a Slack channel, a WhatsApp group, a newsletter someone forwarded. Treating them as brand-driven direct traffic would credit the wrong thing.
Dark traffic and dark social
Dark social is the best known slice: links shared privately in messaging apps rather than posted publicly. A colleague drops your article into a Slack channel, forty people click it, and every one of those visits arrives with no referrer.
Dark traffic is the wider category. It includes dark social plus every other route that loses the referrer:
- Email clients, both desktop and mobile, which almost never send one.
- In-app browsers inside social and news apps.
- Documents: PDFs, slide decks, spreadsheets shared internally.
- Sites setting a strict Referrer-Policy, which trims the header to the origin or removes it.
- HTTPS to HTTP navigation, where the referrer is dropped by the browser.
- Private and incognito windows in some configurations.
The share has grown steadily as messaging replaced public posting and as browsers tightened referrer policies by default. For many B2B sites, private sharing now moves more traffic than any public social feed.
Making dark traffic visible
Estimate it from landing pages
Direct sessions landing on a long, hyphenated deep URL were not typed. That share of the direct bucket is your working dark-traffic estimate.
Tag everything you send
Anything you control, from newsletters to conference QR codes, should carry a UTM. That converts dark traffic into attributed traffic at the source.
Give people a taggable link to share
A copy-link button that appends a lightweight utm_source=share tag turns private sharing into a measurable channel without tracking anyone.
Ask on signup
A how did you hear about us field is the only instrument that sees a podcast mention or a recommendation from a friend. It is unfashionable and it works.
Live demo
See it on real data. The live demo is a fully populated dashboard on demo traffic, with no signup and no card.
Open the live demoWhat dark traffic means for measurement
Dark traffic is not a bug in your analytics and no tool eliminates it, because the information genuinely never reaches your server. Any vendor claiming to unmask it is either fingerprinting or guessing.
The practical consequence is that channel reports systematically undercount word of mouth and overcount whatever is easy to tag. If a campaign looks flat but signups rose, dark traffic is a likely explanation, and the self-reported source field is how you confirm it.
mrkr's position on this is deliberate: it reports direct traffic honestly as unattributed rather than inventing a source for it, and it gives you the landing-page breakdown you need to size the bucket yourself. Guessing an origin would make the report look better and the decisions worse.
FAQ
What is the difference between dark traffic and direct traffic?
Direct traffic is the reported bucket: every session with no referrer and no campaign tag. Dark traffic is the part of that bucket which came from a real source that was lost in transit, as opposed to someone genuinely typing your URL or using a bookmark.
How much of my traffic is dark?
Estimate it from where direct sessions land. Anything arriving directly on a long article slug or a deep page was shared, not typed. On content-heavy sites that regularly accounts for half to two thirds of the direct bucket.
Can analytics reveal the real source of dark traffic?
No, and be sceptical of tools that claim otherwise. The referrer was never sent, so the information does not exist on your server. You can only reduce dark traffic by tagging outbound links, or estimate it from landing-page patterns and self-reported sources.
Start with one line of HTML.Your first visitors show up within a minute.