Definition
Dark traffic is the visits your analytics files as direct because the referrer never arrived. Unlike AI search traffic that keeps its referrer and appears under an assistant’s domain, dark traffic has no origin attached, so a visit caused by a ChatGPT answer looks identical to a bookmark. The cause is invisible, the visit is not.
At a glance
- CategoriesMeasurement
- Related fieldsAnalytics
- DifficultyIntermediate
At a glance
- Dark traffic sits inside the direct channel alongside bookmarks, typed URLs and untagged email.
- Stripped referrers come from desktop apps, in-app browsers, copied links and some privacy settings.
- A visit that starts with a brand name read in an AI answer arrives as direct even days later.
Why the referrer goes missing
A referrer is a header the browser sends saying where the click came from. When it arrives, analytics reads it and files the session under that source.
Ordinary things stop it arriving. Native apps frequently send nothing. Some in-app browsers report the app rather than the page. A user who copies a URL into a fresh tab is not clicking at all.
AI assistants make this worse for a reason unrelated to headers. A shopper reads your brand name in an answer, does not click, and returns days later by typing it. No technology could attribute that visit.
Server logs recover some of what analytics lost, since log file analysis records requests that never ran a tracking script.
Good work that reports as nothing
The pattern repeats. Someone spends a quarter getting cited in answers, opens the channel report, sees AI search traffic flat, and concludes the work failed.
Direct traffic rose over the same period. Branded search rose. Nobody joined the lines, because they live on different tabs.
The cost is not the missing number. It is the decision that follows: a working program gets cut on evidence that was never able to show it working.
The fix is boring. Record a baseline before you start, watch direct and branded search as a pair, and read them against the months your citation rate moved.
Shrink the dark share with what you already run
- Split direct into named sub-buckets. Pull known bookmarks and internal traffic out so the residual means something.
- Baseline direct traffic per landing page. A rise on one deep page is a far stronger signal than a rise on the homepage.
- Tag everything you control. Email, social, partner links, PDFs. Every untagged campaign makes the dark bucket noisier.
- Add a source question at checkout or signup. Self-reported origin is the only method that catches a buyer who never clicked.
- Read server logs monthly. They show assistant crawlers fetching a page, which often precedes the direct traffic to it.
What lands in direct, and what you can do about it
| Source of the visit | Recoverable? | Method |
|---|---|---|
| Assistant link with referrer intact | Yes | Channel grouping on the assistant domain |
| Assistant link opened in a desktop app | No | Baseline comparison only |
| Brand name read, typed in later | No | Self-reported survey at checkout |
| Untagged email or partner link | Yes | Add campaign tags |
| Genuine bookmark or repeat visitor | Not needed | Separate returning visitors out |
Frequently asked questions
-
Is dark traffic the same as direct traffic?
No. Direct is the bucket, dark traffic is the part of that bucket whose real origin was lost. Bookmarks and typed URLs are direct and perfectly explained. Dark traffic is what remains once you take those out, and it is the part that moves when your visibility in AI answers changes.
-
Can I ever measure dark traffic exactly?
No, and any tool claiming an exact figure is modeling, not measuring. What you can do is bound it: baseline the direct channel, tag everything you control, ask buyers directly, and read the residual as a range. An honest range beats a precise number nobody can reproduce.
-
Is dark traffic a new problem?
Not at all. Untagged email, messaging apps and copied links have filled the direct bucket for years. AI assistants add volume and one new twist: the delay between reading your name and visiting your site. The habits that fixed the old version fix most of the new one.
-
Should I report dark traffic to my leadership team?
Report the method, not a headline number. Say what sits in direct, what you removed, how the residual moved and what changed in the same window. A leadership team will accept an honest range with reasoning attached. They will not forgive a confident figure that collapses under one question.