Dark Social · Dark Funnel
Dark Social / Dark Funnel
What is dark social?
- Untraceable sharing and spread via KakaoTalk, DMs, email, etc.
- The dark funnel: invisible in your metrics, yet decisive in buying decisions
- Ask directly to find the real awareness path
The first thing that changes once you accept dark social is how you evaluate channels. The old habit of writing off traffic logged as 'direct visits' as an underperforming channel and cutting its budget first no longer holds up. Activities with unimpressive click counts, such as community engagement or newsletters, are often generating leads, while the dashboard credits search ads. Even when a prospect says in a sales meeting, 'I reached out after seeing material a colleague sent me,' that record appears nowhere in the system. The conversation can only start once the team agrees that what your measurement tools fail to capture is not the same as what has no impact.
In practice, it’s better to keep the two terms separate. Dark social is a problem of 'path': you don’t know how a link was passed along. Dark funnel refers to a state in which the entire buying journey, from awareness through evaluation to internal approval, takes place outside your tracking range. It also differs from zero-click, where people get answers in search results and never visit your site, because with dark social the click happens but the source disappears. The self-reported attribution used as an alternative has its own conditions. Multiple-choice options alone push responses toward 'search,' so add a one-line open-ended field, and keep it optional rather than required so that it doesn’t increase form drop-off; only then will data accumulate. It’s safer to store responses verbatim and rework the classification scheme every quarter.
The most common mistake is using dark social as an alibi. Once every weak campaign result is explained away with 'that happens in the dark funnel, so it doesn’t show up,' nothing can be verified, and the team that loses credibility in budget meetings is ultimately marketing. Treating self-reported answers as the final truth is risky too. Respondents usually name their most recent touchpoint, so a relationship that began at a seminar a year ago gets recorded as 'search.' Worst of all is spamming promotional links into group chats and communities because they can’t be tracked. Dark social only works when third parties share on your behalf; the moment a seller barges in and posts links, the result is getting kicked out and losing reputation. We also often see the opposite failure: a team cuts ads entirely, bets everything on community, and watches quarterly leads fall by half.
John KimInstructor John Kim, author of B2B Marketing Design and CEO of Nanum Consulting
Euijung KimBrand strategist with 30 years in advertising, formerly a division head at Daehong CommunicationsThe term 'dark social' was coined in 2012 by Alexis Madrigal of The Atlantic to describe untraceable sharing traffic that moves through email and messaging rather than social media buttons. In B2B marketing, the concept later expanded to the entire buying journey and evolved into the 'dark funnel.'
Consider a hypothetical case: Company A, an HR-management SaaS firm with 70 employees and KRW 14 billion (≈ US$10,474,300) in annual revenue. Of 118 demo requests over the last three months, 71 showed up in its analytics tool as 'direct visits,' leaving it with no way to explain what its ad spend was achieving on a budget of KRW 30 million (≈ US$22,400) a month. When the CEO asked in an executive meeting, 'Why do we keep spending when we don’t even know where leads come from?', a six-month cleanup effort began.
- When Marketing Manager Kim dug into the traffic logs, half of the 71 'direct visits' turned out to be searches for the exact company name. Kim concluded that 'someone told them our name first.'
- At the bottom of the inquiry form, we added 'How did you hear about us?' as an optional open-ended field. Over two months the response rate was 64%, and 76 responses came in.
- Of 76 responses, 41 were 'recommended in an HR managers’ group chat' or 'a former colleague sent me the link,' and only 3 responses mentioned the company blog directly.
- When Sales Manager Lee changed the first question of every first meeting to 'Where did you first hear about us?' and logged the answers in CRM notes, 9 deals turned out to have started with a webinar six months earlier.
- After a quarterly HR trend report was released as a PDF with no login and a copy-link button was added, responses mentioning group-chat sharing rose from 11 to 29 in two months.
- Six months later, monthly inquiries rose from 40 to 58 and the source-identification rate from 22% to 71%. Manager Kim moved KRW 12 million (≈ US$8,980) of ad spend into report production and community management.