Ahrefs: 97% of llms.txt files got no requests in a month
Ahrefs looked at who actually requests llms.txt. Across 38,360 sites with a valid file, 97% received no requests at all in May 2026, from bots or people.
Last updated
What happened?
Ahrefs published the study on 15 June 2026, written by Louise Linehan with Xibeijia Guan. It used two of Ahrefs’ own data sets: Web Analytics, and Bot Analytics, which reads server logs and live traffic with user agents.
What did Ahrefs measure?
It covered 137,210 domains that had traffic in May 2026, and only the root /llms.txt file. To count, a file had to return HTTP 200 and be real Markdown, not an HTML error page.
| Finding | Figure |
|---|---|
| Domains with a valid llms.txt | 28% (38,360), which Ahrefs calls an upper bound because its customers skew technical |
| Files with no requests in May 2026 | 97%, from any bot or person |
| Where the requests went | About 1,100 domains (3%) received all of them, roughly 22,000 requests |
| Who made the requests | 96% bots, 4% people |
| Requests from named AI bots | 19.5% |
| Requests from GEO and AEO scanners and researchers | About 12% |
In its own words: “Of the ~38,000 domains with a valid file, 97% saw no requests for it whatsoever in May.”
What does it mean for your site?
Most llms.txt files are not being read yet. Don’t expect traffic or visibility from publishing one.
That is an argument for keeping the file cheap, not for skipping it: a small, accurate file costs minutes to make and is ready for the readers that do request it, which so far are mostly coding tools and documentation platforms. See Do I need llms.txt?
Sources
Source: Ahrefs: llms.txt study
Coverage: Search Engine Journal, 16 June 2026