AI citations favor listicles, articles, product pages: Study

AI search citations favor a small set of codecs. Listicles, articles, and product pages drive over half of all mentions throughout main LLMs, in response to new Wix Studio AI Search Lab analysis analyzing 75,000 AI solutions and greater than 1 million citations throughout ChatGPT, Google AI Mode, and Perplexity.

The findings. Listicles led at 21.9% of citations, adopted by articles (16.7%) and product pages (13.7%). Collectively, these three codecs made up 52% of all AI citations.

• Articles dominated informational queries, cited 2.7x greater than different codecs.
• Listicles captured 40% of commercial-intent citations, almost double every other kind.

Why intent wins. Question intent — not trade or mannequin — most strongly predicts which content material will get cited. This sample held throughout industries, from SaaS to well being.

• Informational queries skewed closely towards articles (45.5%) and listicles (21.7%).
• Industrial queries had been led by listicles (40.9%).
• Transactional and navigational queries favored product and class pages (round 40% mixed).

Why we care. This analysis signifies that you simply need to map content material sorts to consumer objectives quite than simply creating extra content material. Articles educate, listicles drive comparability, and product pages convert. Aligning content material format with consumer intent might assist you seize extra AI citations and improve visibility.

Not all listicles carry out equally. Third-party listicles accounted for 80.9% of citations in skilled companies, in comparison with 19.1% for self-promotional lists. That appears to point LLMs desire impartial, editorial comparisons over brand-led rankings.

Mannequin variations. All fashions favored listicles, however diverged after that.

• ChatGPT leaned closely into articles and informational content material.
• Google AI Mode confirmed essentially the most balanced distribution.
• Perplexity stood out, with 17% of citations coming from discussions like Reddit and boards.

Trade patterns. Content material preferences shifted barely by vertical:

• SaaS {and professional} companies over-indexed on listicles.
• Well being favored authoritative articles.
• Ecommerce unfold citations throughout listicles, articles, and class pages.
• Residence restore confirmed essentially the most even distribution throughout codecs.

The analysis. The content types most cited by LLMs