Dark social attribution for B2B teams

What is dark social?

Dark social is the sharing of links and content in places analytics cannot see: Slack and Teams channels, email threads, messaging apps and private communities. In B2B this is where buying groups pass around articles, comparisons and vendor pages before anyone fills in a form.

The sharing itself is not the problem. The problem is that the visits it creates look like nothing in particular, and the content that started them gets no credit.

Why does it show up as direct traffic?

Analytics tools work out where a visit came from using the referrer and any campaign parameters on the link. When both are missing, Google Analytics puts the session in the Direct channel: source exactly matches (direct) and medium is (none) or (not set).^[1] Google describes Direct as users arriving via a saved link or by typing your URL.^[1]

Private shares lose that context in several ways:

• Apps do not always send a referrer. A link opened from a desktop app or a mail client often arrives with no referring page.

• Browsers trim what they send. The default referrer policy sends only the origin, not the full page, on cross-site requests, and nothing when going from HTTPS to HTTP.^[2] Sites can also set a policy that sends no referrer at all.^[2]

• People share clean links. Someone copies the address bar or a link from a document, without the tags your campaign added.

• Browsers strip some parameters. In Safari’s Private Browsing, a subset of query parameters identified as being used for cross-site tracking is removed, while parameters used only for campaign attribution pass through.^[4]

Can UTMs fix it?

Only for links you send. Campaign parameters tell Google Analytics the source, medium and campaign of a visit, and Google says to always set those three.^[3] If a parameter is missing, reports show (not set).^[3]

So tag every link you control: newsletters, social posts, partner placements and sales email templates. Use lowercase values, because Google Analytics treats them as case sensitive.^[3] Our guide to campaign taxonomy mapping covers naming at scale.

What UTMs cannot do is follow a link once a buyer copies a clean version and pastes it into a team channel. For that, you need signals that do not depend on the link.

How does self-reported attribution help?

It asks the buyer directly. A “How did you hear about us?” field on your booking or contact form is the most direct dark social signal you can collect, because it captures the moment no tag sees.

Keep it simple:

• One required question, placed near the end of a high-intent form, not on every download.

• A short list of options plus a free-text “Other”.

• Options that name private sharing in plain words, such as a colleague, a community or a podcast.

Free-text answers are messy but often the most useful. Review them monthly and map them to a small set of categories in the CRM.

What can sales add?

Context the form misses. In discovery calls, ask one standard question, such as what the buyer read or heard before reaching out, and who shared it. Log the answer in a structured CRM field, not only in call notes.

This works best when sales and marketing agree on the categories in advance. A field that each rep fills differently cannot be reported on.

How do you read direct traffic?

Split it rather than ignore it. Direct visits to your homepage are often people who already know you. Direct visits that land on a deep page, such as a specific article or comparison, with no campaign tag, are more likely to have come from a shared link.

Useful views include direct sessions by landing page, direct sessions to deep pages over time, and the share of booked meetings whose first known visit was direct. Treat any “inferred dark social” segment as a direction, not a count. It mixes private shares with bookmarks and returning visitors.

How should B2B teams use these signals?

Combine them, and use them to make content decisions rather than to claim precise credit:

• Compare self-reported answers with tracked sources to see which channels buyers name but analytics underrates.

• Watch which pages attract direct visits to deep URLs, and make those easy to share and quote.

• Support the people buyers already trust, such as your own experts, as covered in B2B employee advocacy.

For budget questions, pair these signals with aggregate methods such as holdout tests and mix models. MTA vs MMM explains when each fits, and closed-loop attribution shows how to keep the CRM as the record.

Where should a team start?

1. Add a self-reported attribution field to your main booking form.

2. Add a standard discovery question and a matching CRM field.

3. Tag every link you send with source, medium and campaign.

4. Build a direct traffic report by landing page and review it monthly.

5. Compare what buyers say with what analytics shows, every quarter.

Our analytics and attribution service covers these steps as part of our B2B demand generation work.