How to build a per-market source list: see who a market reads before deciding what to fix
The top cited source differs by market — YouTube in Poland, a local business daily in Lithuania — and Chinese platforms differ between app and web.
Key numbers
- Cross-market evidence: 128 brands, 12 home markets, 13 languages and 167,551 URL-grounded citations — Wikipedia was the most-cited domain in 11 of 12 languages, the exception being Lithuanian, where the business daily vz.lt led at 4.38%.
- The marginal differences matter more: for 46 Polish national brands the top source was YouTube, and recruiting portals supplied 637 citations — more than double Polish Wikipedia at 297.
- Sources concentrate heavily: about 18% of domains supply 80% of citations (Zipf distribution, α = 0.86, R² = 0.983).
- Chinese search adds inside-platform divergence: across four platforms and eight web/app interfaces, the same platform's app and web source sets differ systematically.
Method and sample
Synthesis of two empirical studies: “How Large Language Models Source Brand Reputation Across Languages and Markets” (arXiv:2606.25787, 167k citations across 12 markets and 13 languages) and the Chinese generative search audit (arXiv:2607.15771, eight interfaces × 614 queries).
What it means for brands
You cannot set a market's source list from one global spreadsheet: first see who AI actually cites in that market — which platforms, which language, which surface — then decide what to cover. Getting the order wrong spends budget where AI never looks.
How SourceSight handles it
SourceSight runs one project per target market, builds the off-site list from that market's language and platforms, orders it by who AI already cites, and groups multi-market projects so source structures can be compared across markets.
Limits
The two studies skew to Europe, North America and Chinese-language search, so coverage is not global; citation shares are counts, not the probability that a user sees the brand.
References
Run it on your own project
What the same prompts return across 10 engines is only knowable by running them.
Related research
Same engine, different surface, different answer: test app, web and API separately
On the same platform and the same questions, app and web cite systematically different source sets — and the API shows something else again.
Methods · 2026-10-07How to design a retest you can trust: five moves so noise does not pass as progress
AI visibility is a distribution, not a point: fix the prompts, repeat the runs, add controls, paraphrase, validate by hand — or you cannot attribute the change.