Products Solutions Templates Pricing Resources

Referral spam: identifying and removing fake traffic sources

Home / Everything About / Everything About Analytics / Referral spam: identifying and removing fake traffic sources

Your referral traffic report shows you're getting visitors from "sexyvideos.com" and "free-download-movies.net." You've never heard of these sites. You never built partnerships with them.

This is referral spam. Fake traffic designed to make it look like traffic came from sites that didn't actually refer it. It pollutes your referral report and makes it impossible to see which real sources drive your traffic.

What is referral spam and how does it work

The basic mechanism

Referral spam bots visit your site and claim to come from another website. They forge the referrer header in their request to make it look like the traffic came from Site X. Your analytics records the fake referrer.

The bot never actually has to visit the fake referrer site. It just has to convince your analytics that it did. This makes referral spam cheap to generate at scale.

Why spammers do this

Spammers create referral spam to advertise sites and services. They want their domain appearing in your reports so you'll visit it. They hope you'll think "this site is sending me traffic, maybe I should check it out." Most of the time it's gambling sites, video sites, or sketchy services.

Some referral spam is more malicious. It's designed to pollute your data so you can't trust your reports. Some might probe for security vulnerabilities. Most is just spam.

How to identify referral spam

Look for referral sources that don't make sense

You get referral traffic from sites you've never heard of. Sites you don't partner with. Sites whose content is nothing like yours. Obvious red flags: gambling, adult content, free movie/music sites, file download services.

Legitimate referral traffic usually makes sense in context. A design blog referring you makes sense. Random keywords from sketchy sites doesn't.

Check conversion patterns from referral sources

Real referral traffic converts at reasonable rates. Spam traffic usually converts at zero or near-zero rates. If a referral source brings 500 visitors but zero conversions, it's probably spam.

Look at bounce rate too. Real referrals have normal bounce rates. Spam bots often land, trigger a pageview, and leave instantly. They'll have near-100% bounce rates.

Examine the traffic pattern

Spam traffic follows predictable patterns. All from the same IP or IP range. All to the homepage. All at the same time of day. All with identical session durations. Real traffic is messier and more random.

Check against your server logs

Server logs show the actual referrer header sent by the request. Check if the suspicious referral source actually exists in your logs. If you see fake referrer headers in logs, you've confirmed spam.

Common referral spam sources

Obvious spam domains

These are clearly spam: sexyvideos.com, adult-related domains, gambling sites, free download services, porno sites. If it's obvious what the site is and it's unrelated to your business, it's spam.

Fake social networks

These spoof legitimate social networks: "social-buttons.com" (pretends to be Facebook buttons), "uberant.com" (pretends to be social sharing). They're trying to look legitimate but are designed to manipulate analytics.

Ghost referrals

These don't even request your site. They send traffic to your analytics tracking code directly without making an HTTP request. They're harder to detect because they skip the referrer header entirely.

How to block referral spam

Create a filter by referral hostname

Create a filter that excludes specific referral domains. Google Analytics lets you filter by hostname. Add known spam domains to an exclusion list. Be conservative—only exclude domains you're sure are spam.

Use a blocklist

The Referrer Spam Blocklist (referrerspamlist.com) maintains a database of known spam domains. Many tools integrate with it directly. You can also manually copy spam domains into your exclusions.

Exclude direct traffic that's actually referral spam

Ghost referrals show up as direct traffic. If you see spikes in direct traffic with no corresponding increase in other metrics, you've probably got ghost referral spam. You can't filter it easily, but you can segment it out of your reports.

Filter by hostname validity

Some referral sources have invalid hostnames. A referral that claims to come from a domain that doesn't exist. You can create a filter that excludes referrals from domains that don't resolve to actual websites.

The trade-off: being too aggressive with filters

Over-filtering can hide real traffic

If your referral spam filter is too broad, it might exclude real referral sources. A filter excluding all shortlinks might block bit.ly links from real referral partners. Be specific about what you exclude.

Historical data can't be recovered

Once you apply a filter, historical data is affected. You can't go back and unfold the filtered data. Create segments before applying aggressive filters so you can see both versions.

Document your filters

Keep a record of every filter and why it exists. This prevents someone from accidentally removing a filter thinking it's outdated. It also helps your team understand what referral data actually represents.