"Direct" is the label Google Analytics reaches for when it genuinely doesn't know where a visitor came from. Somewhere along the way — a stripped UTM, a link shared in a private chat, a browser quietly blocking referrer data — the trail went cold, and the visit gets filed under a channel that isn't really a channel at all. When that visitor converts, the lead is real. The story of what actually brought them in is gone.
Marketers have taken to calling this dark social, or more broadly, dark lead spend: the ad budget and pipeline value sitting behind conversions whose true source never made it into your reporting. This post covers where the term comes from, what the research actually says about how big the problem is, and the honest way to estimate it for your own numbers — including the exact formula behind our free attribution health check calculator.
What the research actually shows
Two credible data points are worth knowing, and worth being precise about — a lot of "dark traffic" content online cites vague or unsourced percentages, which defeats the point of an article about trustworthy attribution:
- Normal direct traffic is a fairly narrow band. Ruler Analytics' guidance is that direct traffic should account for roughly 5–20% of a healthy site's overall sessions. A number well outside that range, in either direction, is worth a second look.
- Social platforms vary enormously in what they tell analytics tools. A controlled 2023 study by SparkToro and Really Good Data sent 1,113 tracked visits across 11 social networks and 16 sharing types. The result: 100% of visits from TikTok, Slack, Discord, Mastodon and WhatsApp arrived with no referral information at all — Google Analytics simply files them as direct. Facebook Messenger lost 75% of its referral data, Instagram DMs 30%, public LinkedIn posts 14%, and Pinterest 12%. Public posts on YouTube, Facebook, and X/Twitter, by contrast, mostly came through with proper attribution intact.
Put together, that's the actual shape of the problem: direct traffic isn't inherently suspicious, but a meaningful and unevenly-distributed slice of it is genuinely mislabeled — and which channels get hidden depends heavily on where your audience shares links.
Turn this into a number for your own business. Enter your ad spend, lead count and lead value, and our calculator applies an adjustable version of this logic instantly.
The other half: stripped UTMs, not just dark social
Dark social explains why certain platforms under-report. A separate, more mundane set of causes explains why "direct" is inflated even for traffic that never touched social media at all:
- UTMs get dropped in transit. Redirect chains, link shorteners, and some HTTPS-to-HTTP hops strip query parameters before analytics ever sees them.
- Closed apps don't pass a referrer. A link opened from an email client, a PDF, or a messaging app's in-app browser frequently arrives with no referral information, tagged or not.
- Browser privacy features intervene. Safari's Intelligent Tracking Prevention and similar mechanisms in other browsers are designed to limit cross-site tracking, and one side effect is a higher rate of sessions landing in direct or unassigned buckets.
None of these are analytics platforms being broken. They're doing exactly what they're built to do — report what they can actually observe. The gap is between what's observable and what's true.
Turning this into an honest estimate
Here's the exact logic behind the calculator, so there's no hidden math:
- You provide three real numbers you already know: monthly ad spend, monthly leads, and average lead value.
- You set one assumption: the share of those leads you believe are actually misattributed — a slider defaulting to 20%, at the upper end of Ruler Analytics' normal direct-traffic range, adjustable to whatever your own GA4 direct/(not set) percentage actually shows.
- The calculator multiplies your lead count by that rate to estimate misattributed leads, then by your lead value to get an estimated dark lead value — and separately applies the same rate to your ad spend, to show how much of your budget's real performance is similarly obscured.
That's the whole formula. No hidden multiplier, no invented industry-wide constant presented as fact. It's designed to give you an order of magnitude — enough to know whether it's worth pulling your own GA4 direct traffic report and checking the real number — not a certified measurement.
What actually closes the gap
A calculator can only ever estimate. Closing the gap for real means capturing the visitor's actual source at arrival and carrying it through to whatever they eventually do — fill a form, book a call, place an order — rather than reconstructing it after the fact from an incomplete session record. That's the specific problem Lead Recorder's lead source tracking is built to solve: the source is attached to the lead itself, so there's no estimate to make.
Ready to replace the estimate with an actual answer? Lead Recorder ties every form fill and call back to its real source automatically.
