An experimental guide to Answer Engine Optimization

(mapledeploy.ca)

8 points | by rosslh 3 days ago ago

3 comments

  • maltelandwehr 3 days ago

    Did you actually conduct the experiment? What was the outcome?

    Based on my experiments (across hundreds of thousands of LLM chats) llms.txt does not have any measurable impact in the vast majority of scenarios.

    • rosslh 3 days ago

      I don't have much in terms of hard data unfortunately. My site is very new, so measuring the AEO impact isn't really possible yet (no baseline to compare against). I am, however, collecting analytics on whether various routes are being hit (e.g. llms.txt and .md routes), as well as the frequency of various AI user agents, so I can do a follow-up post once I have some more data.