Dark Social: What it is and Why Your Analytics Are Lying to You

Dark social includes channels like Slack channels, texts and WhatsApps that your analytics simply can't track. For most brands, it's the largest invisible slice of your word-of-mouth activity.

You can't track it easily. It's a proper commercial blind spot, sitting right in your dashboard.

What is dark social?

Dark social is word-of-mouth sharing that happens in private, untrackable channels like WhatsApp, email, DMs, texts, in-person conversation, and shows up in your analytics as 'direct traffic' instead of the referral it actually was.

Alexis Madrigal coined the term in a 2012 Atlantic article, after noticing a huge chunk of the site's traffic came from sources analytics tools couldn't identify. It landed as "direct." Everyone assumed that meant people typing the URL straight in. The volumes didn't add up. The content being accessed was too specific, too deeply linked, for that explanation to hold.

Why dark social matters more than you realise

Word-of-mouth gets cited again and again as the most trusted, highest-converting channel there is. A friend's recommendation beats any ad, influencer post, or piece of branded content you could commission.

However, most of that word-of-mouth happens inside dark social: private messages, conversations and group chats that you can't track.

The result? You're undervaluing your best customers. Picture the customer who refers three friends over WhatsApp, and all three buy. In your data, they look like a normal customer. The revenue they generated gets attributed to direct, or to whatever channel their friends happened to click last before converting. The advocate gets no credit. Neither does the channel.

That means:

You're underinvesting in advocacy. You can't see the revenue your advocates generate, so you can't justify reaching them properly. Budget flows to paid channels where attribution looks cleaner, even when the economics are worse.

Your LTV models are wrong. A customer's true value includes the revenue from everyone they've referred. Miss that, and you undervalue your best customers. You might even churn them, because they don't look high-value on paper.

Your CAC calculations are incomplete. Paid channels look more efficient than they are, because organic advocacy revenue keeps getting credited to direct or last-click. Real CAC for paid is higher than you think. So is the real cost of losing an advocate.

What dark social traffic looks like

Here's a scenario playing out thousands of times a day for brands with active customer communities.
A customer buys a product they love. They message their friend group on WhatsApp: "Just bought this, you all need to try it, here's the link." Three of the five click through. Two buy within the week.

In your analytics:

• The original customer shows up as a normal purchaser

• The two new customers show up as direct traffic

• No referral gets attributed

• No advocate gets identified

• No word-of-mouth value gets captured anywhere

Now multiply that across your whole customer base. How many WhatsApp shares happened last month? How many texts? How many in-person recommendations? Nobody knows unless you've specifically built to track it.

How to measure dark social

Full visibility will always be out of reach. Private conversations are private by design, and that's fine... nothing here needs fixing at the source. But you can capture far more of it than most brands currently manage.

Name-based attribution. The most effective way to capture dark social referrals at scale. Instead of sharing a link, a customer recommends a brand by sharing their name. The referred friend enters that name at checkout. The referral gets attributed.

The advocate gets identified. This works over WhatsApp ("tell them Dan sent you"), in person, on the phone — anywhere a link is impractical or simply won't get clicked. Name Share® technology is built for exactly this: capturing word-of-mouth that link-based attribution misses by design.

Post-purchase surveys. A simple "how did you hear about us?" question at checkout picks up dark social referrals that links will never catch. It won't get everything, but stacked alongside other attribution mechanisms, it fills real gaps.

It also tells you something about the shape of your word-of-mouth. If 30% of new customers say a friend told them, and your referral programme only attributes 5% of revenue to referral, you've found a measurement gap worth chasing.

Direct traffic analysis. Not all direct traffic is dark social, but a meaningful chunk is. Worth checking: direct traffic that converts at unusually high rates (these visitors already know why they're there); direct traffic landing on specific product or landing pages (someone's been sent a link, not typed one); direct traffic spikes that line up with social or PR activity. None of this is definitive attribution. It helps you size up what you're missing.

UTM discipline. For any link you publish that could get shared — email campaigns, social posts, influencer content — strong UTM tagging gives you a better shot at attributing the origin correctly, even after it's been copied and passed on privately. It won't solve dark social. It shrinks the pile of traffic miscategorised as direct.

Dark social and your referral programme

Most referral programmes are built around links. A customer gets a unique link, shares it with friends, friends click it, referral gets attributed. Simple, in theory.

Here's where it breaks: links don't travel well in dark social. They get shared, but not always clicked. Sometimes the recipient hears about the brand verbally and just searches for it directly. Sometimes the link lands on WhatsApp and gets screenshotted rather than tapped. Sometimes the whole conversation happens face to face, with no link anywhere in sight.

A referral programme measuring link clicks alone is quietly undercounting its own performance. The true reach of your word-of-mouth runs wider than your referral dashboard is telling you.

That's why the sharpest referral programmes combine link-based attribution with name-based attribution. Together, they cover the digital channels where links get clicked and the dark social channels where they don't,  giving you a materially fuller picture of what your advocates are actually generating.

The commercial case for dark social

Dark social carries a revenue number, whether you've gone looking for it or not.

Say a meaningful share of your new customers arrive via word-of-mouth that currently shows up as direct, and you've got no way to identify the advocates behind it or reward them for it. Those advocates go unrecognised. Some will keep advocating anyway. Plenty will stop the moment they realise nothing ever comes of it.

Building the infrastructure to see and measure dark social is a customer retention and acquisition project, full stop. The brands that identify their advocates, understand the value those advocates generate, and reward them properly will run word-of-mouth with the same rigour they already apply to paid search.