Claim and evidence
Publish a llms.txt file summarizing the site for language models, hoping to be cited more.
No dated example yet. Treat this tactic as a hypothesis.
Vendor claim
A tool vendor reports numbers about its own customers or product. Useful, but not independent.
Dead · 2025-11 SE Ranking study of ~300,000 domains found no citation effect (Oct-Nov 2025); Google has not said it is a signal
What the numbers show
No published number for this tactic has passed our word-for-word check yet, so there is no chart.
Why it works
The idea is to give crawlers a curated index. A 300,000-domain study found no relationship with citation frequency.
How to run it
- Add the file to the site root if cheap to do
- Do not expect visibility gains
- Measure citations independently
Ways to run it
Design options to test, not promised results.
- Option ACheap add-on
Add the file only if it takes minutes, and spend the real effort on content and mentions.
- Option BSkip it
Leave the file out and track AI citations directly with the methods you already use.
Dated examples
No dated example yet. Treat this tactic as a hypothesis.
Does it still work?
SE Ranking study of ~300,000 domains found no citation effect (Oct-Nov 2025); Google has not said it is a signal source
Risks
- SE Ranking sells SEO tools; analysis is correlational
- Low effort but no evidence of benefit
Sources
- seranking.com
“Both statistical analysis and machine learning showed no effect of LLMs.txt on how often a domain is cited by LLMs.”
✓ quote checked word-for-word on the page - searchenginejournal.com
“doesn't seem to directly impact AI citation frequency. At least not yet.”
✓ quote checked word-for-word on the page
Change log
- 2026-11-20 — harvest v0.1 (acq-search)