ScaleBun
Skip to article

Traffic sources

webAdministrator

Traffic sources in the Marketing area (Acquisition) — route /marketing/traffic-sources.

Updated Reviewed

Traffic sources lives in the Marketing area of the dashboard, under Acquisition.

At a glance#

Dashboard route/marketing/traffic-sources
AreaMarketing (marketing)
GroupAcquisition
PlatformsWeb apps only. Hidden for mobile apps.

What it does#

Traffic sources is the web-native acquisition view — the replacement for Install sources on web, because a website has sessions rather than installs.

Web sessions are grouped two ways: by classified channel, following your own channel grouping rules, and by source/medium. The data comes from the attribution touchpoints the web SDK records — referrer, UTM parameters and landing page — classified server-side at ingest.

When to use it#

As the first acquisition page for a web product. It answers where traffic comes from with your own taxonomy rather than a generic one, which is what makes the channel numbers usable in a conversation.

Classification happening at ingest matters practically: it means the channel is stored with the session, so filtering other reports by channel is fast and consistent.

Workflow#

  1. Read channel first, source/medium second

    Channel is your own taxonomy and is the level decisions are made at. Source/medium is the raw signal underneath it.

  2. Check that classification matches your intent

    A large "other" or "direct" share usually means rules are missing rather than that traffic is genuinely direct.

  3. Investigate direct traffic sceptically

    Direct is where every lost referrer ends up — app-to-web clicks, some email clients, anything stripping the referrer. It is almost never as large as it appears.

  4. Follow a channel into landing pages

    Landing pages carries the acquisition dimension, so you can read one channel's entry pages and conversions.

Permissions and prerequisites#

Web apps only. Requires the web SDK, which captures referrer, UTM and landing data automatically.

Limits and edge cases#

  • Classification is your rules' output. Changing them changes this report, including retroactively.

  • Referrers are lost more often than people expect. Privacy policies, apps, and some email clients strip them; those sessions become direct.

  • UTM tagging is your responsibility. Untagged paid traffic is indistinguishable from organic referral traffic.

  • Sessions, not people. One person visiting three times is three sessions, potentially attributed differently each time.

Troubleshooting#

Direct is enormous. Missing referrers and untagged campaigns. Tag your links; the referrer loss is not fixable.

A channel is missing. No rule matches its traffic — add one in channel grouping.

Numbers differ from an ad platform. Different definitions and different attribution windows. Compare trends.

Where the data comes from#

Reads the attribution touchpoints the web SDK records (referrer, UTM, landing), classified server-side at ingest through this org’s channel rules.

From the SDK

Served by

  • Dashboard

  • SDK ingestion

Traffic sources · Marketing · Dashboard · ScaleBun