Skip to content

Combosquatting: the brand-plus-keyword disguise

Combosquatting domains keep your brand name intact and add a plausible word: notolens-login.com, notolens-support.net. Nothing is misspelled, which is why typo-based detection, built to generate misspellings, does not find them.

Updated

First measured at scale
Kintis et al., ACM CCS 2017
Study scale
468B DNS records over ~6 years
Lifespan
~60% of abusive ones live 1,000+ days

A combosquat is built by addition, not corruption: the complete brand name plus a plausible keyword. paypal-members.com, youtube-live.com, betterfacebook.com: examples cataloged by Kintis and colleagues, who first measured the technique at scale in a study presented at ACM CCS 2017.

What the research found

Analyzing more than 468 billion DNS records across nearly six years, the study reported three findings:

  • Scale and growth. The authors identified 2.7 million combosquatting domains targeting 268 popular US trademarks, 100 times more prevalent than typosquatting domains, and observed increased combosquatting activity year over year.
  • Longevity. Almost 60% of abusive combosquatting domains remained active for more than 1,000 days. Most of those on public blacklists had been active for several months before they were listed.
  • Abuse range. Phishing, social engineering, affiliate abuse, trademark abuse, and infrastructure for advanced persistent threats all appeared in the data.

The persuasion trick is subtle. A combosquat has no spelling error to spot, and the study notes that brands registering their own trademark-plus-keyword domains train users to treat such names as legitimate. yourbank-security.com reads as a department, not a mistake; the appended word explains the unfamiliar full name away.

Why generators cannot find them

Variant generators, including the lookalike domain generator, enumerate corruptions of the label: dropped letters, swapped vowels, homoglyph characters, bit flips. A combosquat corrupts nothing. The brand appears verbatim, surrounded by a word the generator cannot predict, because the attacker chose it for persuasion value rather than typo proximity.

Detection therefore works the other direction: instead of generating what attackers might register, match the intact brand token inside whatever was registered. notolens’s keyword matching looks for the brand in full inside each newly registered label: notolens-login.com is caught whether the added word is login, verify, or something new. Guards limit noise: a brand that only appears inside an ordinary dictionary word is not treated the same way, and brands built on common words are matched on their own extension. The domain watch report shows which combinations were registered against your brand in the last 30 days.

What to do when one appears

Combosquats that serve phishing pages, fake support flows, or copied content are straightforward abuse reports: the registrar and hosting provider each control a layer that can take the content or the name offline. Where the name itself matters, as with a persistent phishing brand or a domain you want recovered, the intact brand token matters in a UDRP complaint: under the WIPO Overview, adding other terms to a recognizable mark does not prevent a finding of confusing similarity.

Sources

Frequently asked questions

Related tools and resources

Ongoing brand monitoring

notolens checks new domain registrations across 1,570 TLDs and app store results daily, search ads multiple times a day, and trademark registers weekly. When a lookalike domain, conflicting mark, copycat app, or ad using your name appears, notolens checks it, explains the risk, and hands you the records and possible next steps.