What the evidence supports
- Three of five real estate homepages returned 403, 405 or 429 to the disclosed research user agent; those records were not scored for missing on-page features.
- Both successful responses had meta descriptions and canonical tags; one exposed Organization-like schema.
- Neither successful site explicitly named a tracked AI crawler in robots.txt and one served a valid llms.txt file.
- The saved SE Ranking gap contained several real estate SEO terms from 110 to 320 monthly US searches, but AEO Engine ranked mostly outside the top 35 for them.
What the real estate sample showed
Presence checks were regex-based against the fetched public response. A missing observation can mean absent markup, client rendering, bot blocking or a response variant, so raw records and hashes are published.
| Observed signal | Count | Practical interpretation |
|---|---|---|
| Successful homepage | 2 of 5 | The feature denominator is two; failed responses remain visible as access observations. |
| Blocked or non-success | 3 of 5 | Bot response behavior is part of the result, not proof of general crawl failure. |
| Meta and canonical | 2 of 2 | Both successfully fetched homepages exposed basic signals. |
| Organization-like schema | 1 of 2 | The sample is too small for an industry adoption claim. |
| Valid llms.txt | 1 of 2 | One valid file was observable among successful responses. |
A better real estate search program
Make location pages substantive
Publish market method, inventory context, services, local expertise and useful next steps.
Stabilize changing inventory
Handle sold, removed and duplicate listings without creating dead-end or misleading index pages.
Publish market evidence
Explain source, date, geography and limitations for every pricing or trend statistic.
Methodology
- Selected five recognizable public sites in each of five industries before fetching results.
- Requested the public homepage, robots.txt and llms.txt with a disclosed RadiusRankResearchBot user agent and a 12-second timeout.
- Recorded final URL, status, title, meta description, canonical, JSON-LD type presence, robots signals, explicit AI crawler agents and valid llms.txt presence.
- Kept blocked, rate-limited and failed fetches in the raw dataset. No failed record was imputed as a missing on-page feature.
Limitations
- Only two homepage responses were successfully analyzed, so feature rates are especially unstable.
- The benchmark did not test listing feeds, map rankings, local packs or individual agent pages.
- The sample contains five named sites and is not statistically representative.
- Regex presence checks do not score accuracy, completeness, content quality, rankings or commercial results.
- A public response can vary by location, user agent, cookies, JavaScript and edge configuration.
Questions
Does this benchmark identify the best real estate website?
No. It records a narrow set of public technical signals. It does not test rankings, usability, content quality, conversion or business performance.
Why keep blocked sites in the sample?
Removing them after collection would bias the result. They remain as failed observations and are excluded from successful-homepage feature denominators.
Sources and data
- RadiusRank 25-site AI readiness sample: Raw public fetch records, response hashes, methodology and five-industry summary.
- Google Organization structured data: Official guidance for organization identity and administrative details.
Found a changed price, product limit or source? Email team@radiusrank.com with the official URL. We preserve the snapshot date instead of silently rewriting historical data.