Quick answer
A referral source is the external page, site or app a visitor was on just before arriving at your site, as reported by the browser's referrer. It is not the same as a channel, which is a bucket your analytics tool assigns, and it is often missing.
Formula: Capture rate = sessions credited to the partner ÷ clicks the partner reports
What a referral source is, and what it is not
A referral source is the external page, site or app that a visitor was on immediately before they landed on your site. If someone clicks a link in a blog post on partner.example and arrives at your pricing page, partner.example is the referral source of that visit. In web analytics the raw value comes from the browser's Referer header (spelled that way in the HTTP specification) or from document.referrer in JavaScript, so it is a fact reported by the visitor's browser, not something you set.
Three things are easy to mix up with it:
- A referrer is the same raw value seen from the technical side: the URL the browser reports. This site already has a page on the referrer. This page is about how to read the value in reports.
- A channel is a bucket your analytics tool puts the visit into (organic search, paid search, email, social, referral, direct). Many channels are decided by the referral source, but not all of them.
- A UTM tag is a label you write into the link yourself. See UTM. It overrides the referrer when you want a visit credited to a campaign.
"Referral source" is also used in healthcare and sales to mean the person or office that sent a patient or client. That is a different thing. Here it only means the page a web visit came from.
Referral source vs channel vs UTM source
| Term | Who sets it | Example value | Can be missing? | Use it to answer |
|---|---|---|---|---|
| Referral source (referrer host) | The visitor's browser | partner.example | Yes, often | Which specific site sent this visit? |
| Channel | The analytics tool's rules | Referral, Organic Social, Direct | No, every visit gets one | How does traffic compare by type? |
| UTM source / medium | You, in the link | newsletter / email | Only if you forgot to tag | Which campaign should get credit? |
Click ID (gclid etc.) | The ad platform | A long token | Only on tagged ad clicks | Was this an ad click? |
The practical rule: use the referral source to find out who sent traffic you did not plan, use UTM to measure traffic you did plan, and use the channel to compare totals. Google's own documentation defines Referral as non-ad links on other sites or apps, and Direct as a saved link or a typed URL, which shows why a visit with no referrer ends up in Direct.
Why the referral source is often missing
A visit with no referral source is not a visit from nowhere. The browser simply did not say where it came from. The usual reasons:
- The link was not in a web page. Clicks from desktop email clients, PDFs, messaging apps and some mobile apps often carry no referrer.
- A referrer policy removed it. Browsers default to
strict-origin-when-cross-origin: cross-origin requests send only the origin (the domain, not the page path), and an HTTPS page linking to an HTTP page sends nothing. A site can also setno-referrerorrel="noreferrer"on its links to send nothing at all. - A redirect in the middle. Link shorteners and some redirect chains can replace or drop the original referrer.
- The visit was not a click. Typed addresses, bookmarks and a browser restoring a tab have no referrer.
- The script ran late or not at all. If the visitor leaves before the tracking script loads, or blocks it, the visit is never counted.
Missing referrers are why Direct is often a catch-all bucket rather than a measure of brand strength. The blog post on direct traffic goes through how to shrink it, and dark social covers the shared-in-private-chat case.
A worked example: where did the clicks go?
Sample data: a partner site reports 1,000 clicks on a link to your pricing page during a month. Your analytics shows 820 sessions credited to the referral channel for that partner. Your site had 12,000 sessions in total.
| Where the 1,000 clicks ended up | Sessions |
|---|---|
| Counted as referral from the partner | 820 |
| Arrived with no referrer, counted as Direct | 110 |
| Visitor left before the script loaded, never counted | 40 |
| Partner had added its own UTM tags, so credited to a campaign instead | 30 |
| Total | 1,000 |
Check: 820 + 110 + 40 + 30 = 1,000.
Two numbers come out of this:
- Capture rate = sessions credited to the partner ÷ clicks the partner reports = 820 ÷ 1,000 = 82%. The other 18% is not lost traffic; it is traffic filed under other labels or never counted.
- Referral share = sessions from the partner ÷ all sessions = 820 ÷ 12,000 = 6.8% by the referral label. The partner's true contribution is (820 + 110 + 30) ÷ 12,000 = 8.0% if you assume the no-referrer and re-tagged visits also came from the partner, which you can only confirm from the partner's own click report.
The decision rule: if the capture rate is high, believe the referral report. If it is far below what the partner reports, ask them to add UTM tags to the link instead of correcting the report by hand.
Common mistakes when reading referral sources
- Treating referral as a channel and a source at once. The referring host is the source; Referral is the channel. A visit from
news.exampleis sourcenews.example, channel Referral. A visit from a search engine has a referring host too, but its channel is Organic Search. - Counting your own site. A click from one of your pages to another is not acquisition. Payment pages and sign-in providers that send the visitor back to you create the same problem; the page is covered in self-referral.
- Reading the host without the page. Under the default referrer policy, a cross-origin referrer is only the domain. You know the partner site, not which article the visitor read.
- Comparing tagged and untagged traffic as if they were the same. Tagged visits are credited exactly; untagged ones depend on the browser. A campaign that moved from untagged to tagged links will look like a jump.
- Ignoring spam. Some "referrals" are bots visiting your site to leave a link behind. See referral spam, and keep bots out of the denominator.
For the wider picture of how sites send traffic and how to grow it, see referral traffic.
Reading referral sources in TapCub
Take the example above and run it in TapCub.
- Open Web analytics › Sources and click the Referrers tab. Rows are referring hosts with visitors, sessions, pageviews, bounce and duration. Find
partner.exampleand compare its sessions with the clicks the partner reported. - Switch between First source of session (the default, each session credited once) and Every arrival, which also counts an external arrival in the middle of a session. With the second setting one visitor can appear under several sources, so rows can add up to more than the total. Use the first for acquisition and the second for how many sites touch your visitors.
- Open the Channels tab and compare. Rules look at UTM, the referring host and click IDs, and rules you add for the site win over the defaults. A visit from your own site is marked internal, and a visit with no source at all is direct.
- Click a row to add it as a filter, then check the UTM tab for the same partner. If the partner added tags, those visits sit under campaign channels, which is the 30 sessions in the table.
- Add a goal and read the conversion column on the Channels tab, so the partner is judged by conversions rather than visits.
TapCub does not repair a missing referrer after the fact. It reports what the browser sent, the UTM tags and the click IDs, and groups them by rules you can see and change. Real people and bots are counted separately, so crawlers do not inflate a referral source.
See also: Backlinks: What They Are, Types and How to Track Them.
Sources
- MDN Web Docs, Referer header.
- MDN Web Docs, Referrer-Policy header.
- Google Analytics Help, Default channel group.
- Google Analytics Help, Traffic-source dimensions.
Was this explanation clear?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.

Product Experience Specialist
Explores usability, user journeys, and feature adoption through practical examples of onboarding and everyday product use.