Who does AI cite about Health in Guadalajara?
Sources Google AI Overviews cites when answering commercial Health queries measured from Guadalajara, from the observatory's open corpus.
The segment's most-cited sources
Based on 8 segment queries that returned a generative answer. · last measured: September 1, 2026
| # | Source | Appears in | Avg. position | Queries |
|---|---|---|---|---|
| 01 | doctoralia.com.mx | 50% | 6.5 | 4/8 |
| 02 | cucs.udg.mx | 13% | 1.0 | 1/8 |
| 03 | riodm.com | 13% | 1.0 | 1/8 |
| 04 | chopo.com.mx | 13% | 2.0 | 1/8 |
| 05 | laserocular.com.mx | 13% | 2.0 | 1/8 |
| 06 | physiohealth.com.mx | 13% | 3.0 | 1/8 |
| 07 | mx.linkedin.com | 13% | 5.0 | 1/8 |
| 08 | analisisclinicos.mx | 13% | 7.0 | 1/8 |
| 09 | dentaldepartures.com | 13% | 21.0 | 1/8 |
How this was measured
- The corpus is a set of commercial Spanish queries we define and publish, versioned in the codebase. It contains no brand names, ours or anyone else's: a query with our name in it would manufacture a citation in our favour.
- Each query is sent once to Google AI Overviews in Spanish. National queries are measured with México as the market; sector-and-city queries with that city's location (Mexico City, Guadalajara or Monterrey), using the same template queries across all three so segments stay comparable.
- We record the sources the engine cites. When the engine wraps a source's link and the exact domain is unrecoverable, we count it by the publisher name exactly as the engine displays it — those rows are marked with ≈. Ignoring them would bias the ranking toward the sources the engine chooses to link directly.
- A source counts once per query. If one answer cites it three times, that is still one appearance — otherwise we would be rewarding verbosity, not visibility.
- The percentage is calculated over queries that returned a generative answer, not over the whole corpus: dividing by the total would understate every figure by the same silent factor.
- A segment is published only if it clears a quality gate (enough distinct sources, none dominating it). Whatever fails the gate is not published.
- No data comes from AEON42 customers. Not their sites, not their queries, not their Search Console.
corpus 2026-08-31.1 · AI_OVERVIEWS
What these numbers don't say
Generative answers are probabilistic: the same query can return different sources across two runs, and the engine changes without notice. This is a snapshot of one corpus on one date, not a stable measurement of the market. It also doesn't cover ChatGPT or Perplexity — only Google AI Overviews. We say so because the alternative is selling certainty nobody has.
Does your site show up?
Check for free whether Google AI Overviews cites you for the search that matters — or connect your Search Console and watch it weekly alongside your real traffic.