Dark social is traffic that came from social media but shows up in your analytics as "Direct", because it was shared through DMs, messaging apps, email or a copy-pasted link with no referrer attached. If your social ROI looks too low, this is the first leak to check.
There is no honest universal number for how big it is. It depends on your audience and your content. This five-step audit gives you your own estimate.
TLDR: Look at Direct traffic to deep pages, line it up with your posting times, tag every link you control, ask people how they found you, and give each channel its own trackable link. The result is a defensible range, not a guess.
Why dark social happens
Plenty of social sharing happens out of sight: someone sends your post to a friend in a chat, a colleague forwards a link by email, or an app opens links in a way that drops the source. In each case the visit arrives with no label, and analytics tools file it as Direct.
Good content tends to get shared privately more than average content, so the better your posts are, the more of their effect hides here.
Step 1: Filter Direct traffic to deep pages
Nobody types a long blog URL from memory. Open your analytics, filter Direct visits, and look only at landing pages that are not your homepage or your main product page.
Deep-page Direct visits are very likely shared links. That is your raw dark social pool.
Step 2: Line it up with your posting times
Plot Direct visits to those deep pages by day, and mark the days you published or ran a campaign. If the visits climb in the day or two after a post and fall back afterwards, social is driving them.
Compare to a week with no posts. The difference is your first estimate.
Step 3: Tag everything you control
Much "dark" traffic is your own untagged link. Fix that first.
- Use one naming convention for campaign tags and keep it in a shared sheet.
- Give each channel its own link, including the one in your bio.
- Use a link-in-bio page so one trackable link covers every post. Mydrop's link-in-bio pages give each profile one place to point.
For the sheet-based version of this setup, see our guide to a simple attribution dashboard in Google Sheets.
Step 4: Ask people how they found you
Add a "How did you hear about us?" field to your signup and checkout forms, as open text rather than a dropdown. People will write "a friend sent me your post" or "saw you on Instagram", which is exactly what the referrer data lacks.
Count the answers monthly. Even 30 responses will show you which channels are under-credited.
Step 5: Score your leak
Use this three-point scorecard each quarter.
| Check | Leaky | Healthy |
|---|---|---|
| Share of Direct visits landing on deep pages | High and rising | Low and stable |
| Direct spikes after posts | Clear spikes | No pattern |
| Self-reported "social" answers vs analytics-reported social | Far higher in forms | Roughly similar |
If two of three read "leaky", report social results as a range: tracked social as the floor, tracked plus the estimated dark portion as the ceiling.
How to report it without overclaiming
State the method, show both numbers, and never present the estimate as fact. A line such as "Tracked social conversions: X. Estimated including dark social: between X and Y" survives a skeptical CFO far better than a single large number.
Our guide on four metrics that prove ROI covers what to put next to it.
Keep your own posts comparable
Dark social measurement is only useful if you can say which posts drove the spikes. Mydrop's analytics show each post against that account's own median, so a spike lines up with a specific post instead of a vague "social".
Next step: do Steps 1 and 2 today. They take half an hour, need no new tools, and tell you whether dark social is a rounding error or a real share of your results. If it is real, set up a link-in-bio page in Mydrop and start tagging.
Google review
Trustpilot review
Product Hunt comment
