Wikipedia holds the top spot with a 3.4% citation share, yet 74% of the most-cited domains can be influenced through marketing. Top-tier brands earn 10 times as many citations as those at the bottom, and the gap is widening.
Between October 2025 and March 2026, Goodie AI analyzed 58.6 million AI citations from four platforms: ChatGPT, Gemini, Claude, and Perplexity. The biggest change from the previous study is that 13 of the top 20 domains were replaced. In the world of AI citations, the ground can shift in just six months.
Wikipedia kept the top spot with a 3.4% citation share (up from 2.2% in the previous study). YouTube and Reddit follow in second and third place. Among the notable changes, eight major news outlets dropped out of the top 20, while five gaming and tech review sites made the list.
| Rank | Domain | Citation share |
|---|---|---|
| 1 | Wikipedia | 3.4% |
| 2 | YouTube | 2.1% |
| 3 | 1.4% | |
| 4–6 | Google, LinkedIn, Alibaba | — |
| 7–10 | TechRadar, Forbes, Tom's Guide, X | — |
| 11–15 | NerdWallet, Instagram, PCGamer, Facebook, Yahoo Finance | — |
Citation rankings vary widely by industry. In retail banking, NerdWallet leads with a 3.9% citation share. In B2B SaaS (CRM), TechRadar posts a remarkable 8.86%. In skincare and cosmetics, a brand called Kaja Beauty takes second place with 4.20%, behind only Wikipedia (4.4%).
The takeaway: The types of domains most likely to be cited differ completely by industry and category. It is essential to map citations in your own category and pin down where you need to earn them.
One key finding is that about 74% of highly cited domains can be influenced through marketing activity. A strategic presence on social platforms, review sites, and commerce platforms is a realistic way to increase your share of AI citations.
Citation share compounds: top-quartile brands earn more than 10 times as many AI citations as those in the bottom quartile. Brands that build citation share early create a long-term advantage, much as they would with domain authority. Moving now lets you open up a wide lead over brands that enter late.
In the global data, LinkedIn, Reddit, and YouTube drive citations. The Japanese market may have citation patterns that differ from those of English-language platforms. Even so, global models such as ChatGPT, Gemini, and Claude also cite Japanese-language content widely. Creating structured content in Japanese, and publishing on LinkedIn, YouTube, and Reddit in both Japanese and English, is the strategy for maximizing AI citations both in Japan and overseas.
FREE TRIAL
Check Your AISEO Score for Free
Just enter your URL to see how visible your brand is in AI search right now.
Get a Free Diagnosis →![How Social Media and Content Formats Shape AI Citations [Research Report]: Inside a 3.8x Rise in Monthly Citations](/img/column/en/social-ai-citation.svg)
