A referrer is the previous-page URL a browser reports in the HTTP Referer header — a spelling frozen by a typo in RFC 1945 in 1996. Analytics tools group those URLs into traffic sources. Since 2020 most browsers trim cross-origin referrers down to the bare origin, so the path and query string never arrive.
Referrer types and what each one actually sends
Five referrer types cover almost all traffic: organic search, paid advertising, social, referral sites and direct. The type is inferred from the domain in the Referer header rather than declared by the visitor, and direct is the residual bucket — every visit whose header arrived empty lands there, whatever the real source was.
| Referrer type | What arrives in the Referer header | How analytics reports it |
|---|---|---|
| Organic search | https://www.google.com/, origin only, no query | A search engine row, usually without keywords |
| Paid advertising | The ad network or search engine origin, with nothing marking the click as paid | Campaign rows, but only from tags on the landing URL |
| Social media | https://t.co/ or https://l.facebook.com/, origin only | One row per network domain |
| Referral site | The linking site’s origin, path stripped | One row per external domain |
| Direct | Nothing at all | Direct traffic, the bucket for missing referrers |
Paid advertising is the one type the header cannot identify on its own, because a paid click and an organic click from the same search engine present an identical origin. Campaign attribution therefore depends on parameters you add to the destination URL, not on the referrer: see tracking UTM and campaign parameters for how those tags are stored and filtered.
Why referrer data goes missing
Referrer data is optional, and four mechanisms delete it. Browsers strip the header on HTTPS-to-HTTP navigation, the default strict-origin-when-cross-origin policy trims the path and query, a site can send Referrer-Policy: no-referrer and suppress the header outright, and apps and email clients often open links with no header at all.
Chrome made strict-origin-when-cross-origin its default in 2020 and Firefox followed in 2021. Google had already withheld search keywords from its own referrer URLs years earlier, and the policy change extended the same trimming to nearly every other site. The loss runs one way — detail disappears and never returns — so a referring domain in a report stays dependable while a full referring URL increasingly does not.
Referrer spam is the opposite failure. A bot forges the Referer header with a URL its operator wants you to visit, and that domain appears in your reports having never sent a real visitor. Because the header is entirely under the sender’s control, filtering is the only remedy: SlimStat’s Referring Sites exclusion accepts wildcards and discards the whole pageview inside the tracker, before any row reaches the database. The exclusions reference gives the field format.
Where referrers appear in SlimStat
SlimStat writes the full referring URL to a referer column on the wp_slim_stats table and reads that column back on a Traffic Sources screen inside the WordPress admin, where the Top Referring Domains report groups the URLs by domain. Both the column and the screen ship in the free plugin, with no external service involved.
Two behaviours are worth knowing before reading those reports. Same-domain referrers are dropped by default, so internal navigation never pollutes the traffic-source list, and a Same-Domain Referrers setting re-enables them for subfolder multisite networks. The tracker also accepts android-app:// referrers alongside http and https, so Google Discover and Android app traffic keeps an identifiable source instead of collapsing into direct. SlimStat has been listed on wordpress.org since 2010 and runs on 70,000+ sites; the Google Analytics alternative page covers how the two tools differ.
Frequently asked questions about referrers
These five answers cover the questions that follow a referrer definition: the difference between referrer and referral, what an HTTP_REFERER value looks like, and why a referrer can be missing. Each answer stands on its own, so no question here depends on the question above it.
What is an example of a referrer?
If someone clicks a link to your site from a post on facebook.com, the referrer recorded is https://l.facebook.com/ or a similar Facebook origin, not the post URL, because cross-origin referrers are trimmed to the origin.
What is the difference between a referral and a referrer?
The referrer is the source URL the browser reports, while a referral is the visit that URL produced. One referrer domain can generate thousands of referrals.
What is an example of an HTTP_REFERER value?
A typical HTTP_REFERER value today is a bare origin such as https://www.google.com/. Full values carrying a path, such as https://example.com/blog/post/, still occur when the linking site sets a permissive Referrer-Policy.
Why is my referrer missing?
A missing referrer means the browser sent no header, which happens on direct visits, bookmarked visits, HTTPS-to-HTTP navigation, links opened from apps or email clients, and any site that sets Referrer-Policy: no-referrer.
Is a referrer the same as a traffic source?
A referrer is one raw URL; a traffic source is the category an analytics tool assigns after grouping referrer domains. Direct traffic is a source with no referrer behind it at all.