What Is a Watering Hole Attack?

A watering hole attack is a technique where an attacker compromises a website that their intended targets are known to visit, then waits for those targets to show up and get infected — rather than attacking them directly. The name comes from predators that stake out a watering hole instead of chasing prey across open ground: it’s easier to wait where the target is guaranteed to come to you.

How the attack unfolds

A watering hole attack typically follows a predictable sequence:

1. Reconnaissance. The attacker identifies a group they want to compromise — employees of a specific company, members of an industry, or users of a niche community — and researches which third-party sites that group routinely visits: an industry forum, a vendor’s support portal, a trade publication.
2. Compromise the site. The attacker breaches that third-party site, usually through an unpatched vulnerability, and plants malicious code — often JavaScript that fingerprints visitors or serves an exploit only to specific browser or OS versions.
3. Wait. The compromised site keeps functioning normally for its regular audience. The attacker doesn’t need to send anything, click-bait anyone, or trigger a spam filter; they just wait for members of the target group to visit as they normally would.
4. Selective delivery. Many watering hole campaigns fingerprint visitors and only serve the exploit to machines matching specific criteria — a particular company’s IP range, a specific browser version, a specific language setting — so the payload stays hidden from security researchers and unrelated visitors crawling the same site.

Why attackers prefer this over phishing

Phishing requires tricking someone into clicking a link or opening an attachment, which means fighting spam filters, security awareness training, and a target who has to make one bad decision. A watering hole attack requires no interaction beyond a target visiting a site they already trust and already visit as part of their normal routine. It also sidesteps email security entirely, since there’s no malicious email to catch. The trade-off is setup cost: the attacker has to find and compromise a legitimate third-party site first, which takes more reconnaissance and technical effort than sending a convincing email.

Watering hole vs phishing vs supply chain attacks

These three techniques are often confused because they all rely on trust the victim already has in something else, but they target trust in different places.

	Watering hole	Phishing	Supply chain attack
What’s compromised	A site the target visits	Nothing — a fake message is sent	Software the target installs or depends on
Trust exploited	Trust in a site’s legitimacy	Trust in a sender’s identity	Trust in a vendor or dependency
Requires victim action	Just a normal visit	Click a link or open a file	Just a normal update or install
Attacker’s setup cost	Compromise a third-party site	Craft a convincing message	Compromise a vendor or package

A watering hole attack is really a specific flavor of supply chain thinking applied to web browsing instead of software dependencies — see our explainer on software supply chain security for how the same logic plays out with packages and build pipelines, and typosquatting for a related technique that exploits trust in a domain or package name.

Detecting and defending against it

From the defender’s side, watering hole attacks are hard to catch early precisely because the delivery site looks and behaves normally to everyone except the intended targets. A few practical defenses reduce the risk:

• Keep browsers and plugins patched. Watering hole payloads usually rely on a known, unpatched vulnerability rather than a novel zero-day — patching promptly closes off most of them.
• Use endpoint detection that flags unusual outbound connections, since a compromised browser process reaching out to an unfamiliar host is a strong signal even if the initial exploit went unnoticed.
• Segment high-value users’ network access so that a single compromised workstation can’t reach sensitive internal systems directly — the same principle behind putting a bastion host in front of production access instead of exposing it broadly.
• Monitor DNS and proxy logs for connections to newly registered or low-reputation domains, which watering hole payloads often use for command-and-control after the initial compromise.
• Content security policies and script allowlisting on your own sites reduce the chance that your organization’s site becomes someone else’s watering hole.

Web application firewalls and WAFs can catch some exploitation attempts against the compromised site itself, but from the victim’s side, the more effective controls are patching, network segmentation, and monitoring outbound traffic rather than trying to detect the compromised site in advance.

The takeaway

A watering hole attack flips the usual phishing model: instead of chasing the target with a convincing message, the attacker compromises a site the target already trusts and waits. That makes it harder to catch with email-focused defenses and easier for the attacker to stay hidden, since only the intended visitors ever see the malicious payload. Patch promptly, segment network access, and watch outbound traffic for the unusual connections a successful compromise leaves behind — those defenses work whether or not you ever spot the compromised site itself.