The llms.txt Effect
Research answer

Does llms.txt increase AI citations?

No measurable lift appears in the study. Median citations are 3 with or without llms.txt, average citations are 6.8 versus 6.7, and the Mann-Whitney p-value is 0.85.

Published 2026-03-18Updated 2026-03-14Last tested 2026-03-14Study 005
[01]

Answer

No. No measurable lift appears in the study. Median citations are 3 with or without llms.txt, average citations are 6.8 versus 6.7, and the Mann-Whitney p-value is 0.85.

The data does not support llms.txt as a reliable citation-growth tactic on its own, meaning teams should not rely on it to drive visibility.

[02]

Evidence

Median citations with llms.txt
3

Median citations for domains with llms.txt.

Median citations without llms.txt
3

Median citations for domains without llms.txt.

Average citations with llms.txt
6.8

Mean citations for domains with llms.txt.

Average citations without llms.txt
6.7

Mean citations for domains without llms.txt.

Median citations with llms.txt
Median citations for domains with llms.txt.
3
Median citations without llms.txt
Median citations for domains without llms.txt.
3
Average citations with llms.txt
Mean citations for domains with llms.txt.
6.8
Average citations without llms.txt
Mean citations for domains without llms.txt.
6.7
Mann-Whitney p-value
No statistically significant citation effect detected.
0.85
[03]

Context

Operators must allocate technical SEO resources efficiently. Knowing that llms.txt does not drive citation growth prevents teams from wasting engineering cycles on a file that yields no measurable performance lift.

What is the median citation difference when using llms.txt?

There is no difference. The median citations are 3 for domains with llms.txt and 3 for domains without it.

Is there any statistical significance to the citation difference?

No. The Mann-Whitney p-value is 0.85, indicating no statistically significant citation effect detected.

What to do next
  • Treat llms.txt as an optional housekeeping file rather than a primary citation-growth lever.
  • Prioritize answer quality, source coverage, and page structure before spending disproportionate effort on llms.txt.
  • Measure discovery and crawl behavior directly if you publish llms.txt instead of assuming it improved citation performance.
[04]

Source & Method

Study
Study 005
Published
2026-03-18
Updated
2026-03-14
Last tested
2026-03-14
Author
Mack Grenfell
Source and methodology

Built from HTTP scans of 37,894 AI-cited domains, linked to 337,362 citations and 882 citation snapshots in the Trakkr corpus.

Flagship study
The llms.txt Effect
Machine-readable data
JSON payload
Citation targets
Source study and JSON stay publicly linked so crawlers can verify the page quickly.
https://trakkr.ai/trakkr-research/llmstxt-effecthttps://trakkr.ai/data/research-answers/llmstxt-effect/answers/does-llms-txt-increase-ai-citations.json
Limits
  • This is an observational study. It measures correlation with citation outcomes, not a controlled experiment.
  • Adoption is uneven by sector, so raw averages can hide category concentration in SaaS and developer tooling.
  • A null citation effect does not mean llms.txt has zero operational value for every workflow. It means the study did not find a measurable citation lift.
Citation block

No. No measurable lift appears in the study. Median citations are 3 with or without llms.txt, average citations are 6.8 versus 6.7, and the Mann-Whitney p-value is 0.85.

[05]

Related