Unanswered

Copilot Studio Unable to Access Data from Certain Websites

(3) Share

Report

Posted on by Suchit Chaudhari

Hi,

I'm using Microsoft Copilot Studio and have integrated my company's case study website to make the content accessible through Copilot.

However, I've noticed inconsistent behavior: Copilot is able to read and fetch data from some case studies, but it fails to access or return data from others—even though all of them are hosted on the same domain and are publicly accessible.

Is there a reason why Copilot might be able to read certain pages but not others on the same site? Are there specific metadata, headers, or page structures that could be affecting this behavior?

Any insights or suggestions would be appreciated!

Thanks!

Categories:

General topics

I have the same question (0)

All responses (4)

Answers (0)

Michael E. Gernaey 53,969 Moderator on at

Like (0)

Report
Copy link

Link copied!

Hi @CU09040310-0

Is there any correlation between at what level in your site the Case Studies are stored for the ones that do not work versus the ones that do?

Was this reply helpful? Yes No
Suchit Chaudhari 17 on at

Like (0)

Report
Copy link

Link copied!

Hi @Michael E. Gernaey,

I've tried adding a specific case study link as a knowledge source and asked questions related to the content, but it seems that Copilot is unable to crawl or retrieve information from specific external websites at all.

Is there a known limitation around Copilot’s ability to access and process data from certain public URLs? Any guidance would be appreciated.

Was this reply helpful? Yes No
Suchit Chaudhari 17 on at

Like (0)

Report
Copy link

Link copied!

Was this reply helpful? Yes No
ronaldwalcott 3,862 Moderator on at

Like (0)

Report
Copy link

Link copied!

There are limitations depending on the type of website. layout, robots.txt and meta tags that can limit interrogation by automated tools.

You should be able to use tools such as Site Scan - Bing Webmaster Tools (this is not created for this purpose) which could help in identifying problems on pages which would prevent them from being read by an automated tool.

Was this reply helpful? Yes No