Continuing my search for practical insights into what actually helps businesses gain visibility in AI search, I came across two new studies this week that reveal something worth paying attention to.
For months, the advice has been “optimise your website for AI.” But the latest data suggests we may need to ask a bigger question:
How many credible places online are telling AI who you are?
Two studies published this month give us some interesting numbers behind that shift.
LinkedIn just entered AI’s top three sources, with an asterisk.
Meltwater analysed approximately 7.3 million citations across eight AI platforms. LinkedIn grew 24.8% in August to 100,200 citations, moving from fifth to third place. YouTube (212,500) and Reddit (127,000) still lead.
The catch is that most of LinkedIn’s citations came from one platform. Perplexity generated 69,302 of them—roughly 69% of LinkedIn’s total in the sample. ChatGPT and Gemini, meanwhile, recorded zero LinkedIn citations in the tracked data.
Each AI platform trusts different sources.
In Meltwater’s data:
- Perplexity leaned heavily on LinkedIn.
- Gemini leaned on YouTube and Reddit.
- ChatGPT and Claude showed more institutional and data-led source profiles, with NIH leading among the report’s top 10 domains for both platforms. Statista was also particularly prominent for Claude.
A strategy built on one channel is a strategy built around one AI platform’s current preferences—and those preferences can change.
Media coverage still matters, but the mix is changing.
Earned/News fell from 33.1% to 29.4% of citation-domain counts, its third straight monthly decline. It still supplied roughly 2.1 million citation-domain counts, close to three in ten.
Meltwater’s advice is essentially to treat strong media coverage as the centre of a wider content system—extending it into LinkedIn posts, videos, FAQs, research pages and other useful content formats.
Your coverage doesn’t have to come only from high-authority publications, with one important exception.
Stacker studied 895 stories that generated about 212,000 media pickups across its network and tracked citations across six AI platforms.
Sites with lower domain ratings (under DR40) still accounted for one-fifth of Stacker network citations. More pickups were also associated with more citations, although for sub-DR40 domains that benefit tended to plateau at around 90 pickups.
The notable exception was ChatGPT.
Only 0.05% of ChatGPT responses in the study cited a pickup from a sub-DR40 site, compared with 0.9% for DR61+ sites. So if ChatGPT visibility is a priority, higher-DR coverage carried substantially more weight in this particular dataset.
Claude, by contrast, cited lower-DR sites at the highest rate among the six platforms studied.
Why it matters for you
Authority isn’t one great article or one great profile.
It’s the same credible story about your business showing up consistently across several places AI systems can potentially find and cite.
The Womenlines Authority Chain
Our framework, not a study finding. It won’t guarantee an AI recommendation, but it builds a network of consistent, independently supported information about you.
Credible media mention ? Your LinkedIn ? Video ? Website FAQ ? Consistent brand facts
This Week’s Action List
- Turn your latest media mention into a LinkedIn post. State the main point early and link back to the original coverage.
- Give your best YouTube video a proper description, accurate transcript and chapters. These provide AI systems with clearer information to interpret. Meltwater specifically recommends detailed descriptions, accurate transcripts and chaptering.
- Add an FAQ to your website answering the questions customers actually ask, with sources where appropriate.
- Aim for a spread of mentions. Include relevant niche and trade outlets, not only the biggest publications. Stacker’s findings suggest both reach and domain authority can matter.
- Check your brand facts. Your name, bio and services should be consistent across your website, LinkedIn, YouTube and media boilerplate.
- Run a mini audit. Ask Perplexity, ChatGPT, Gemini and Claude the questions your customers ask, and note which sources appear. Repeat monthly to watch how the results change.
A note on the data: These findings shouldn’t be treated as universal rules. Meltwater’s study used a defined set of consumer-category prompts, and source preferences can vary by prompt, category, location, timing and AI platform. Stacker’s analysis was specifically based on stories distributed through the Stacker Newswire. Both companies also provide commercial products/services connected to these areas. The studies are therefore best read as useful directional evidence rather than a rulebook for AI visibility.
Sources: Meltwater — AI Search Visibility Report: August 2026 Stacker — What 895 Stories Reveal About the Citations Earned By Sites Under DR60
Charu Mehrotra
Founder Womenlines.com
Editorial Note: Disclaimer: This article is for informational purposes only. Womenlines does not independently verify or guarantee all claims, facts, outcomes, or results mentioned. Readers are encouraged to conduct their own due diligence.
Follow Womenlines on Social Media
