Skip to main content
Solved

How federated search with Zendesk

  • February 7, 2025
  • 2 replies
  • 108 views

Forum|alt.badge.img+3

Hello, my community has set up federated search with our Zendesk knowledge base. Does the searching functionality described in this post generally apply to external resources as well?

Obviously, Zendesk articles don’t have all of the attributes listed here (replies, tags, answers, etc.). Does that leave us with keyword matches in the title, followed by keyword matches in the content? 

 

If there’s any other documentation on this (which I couldn’t find in my own search), I’d love to know!

Best answer by mbuuck1

Updating my post with new information that I learned from ​@Yoeri on the behavior of federated search. He confirmed in this other post that:

The way how federated search works is that we use the API to fetch Zendesk content and add it to the search engine in Gainsight (so you have a ‘bucket’ of content both from Gainsight and Zendesk on which we apply the search). Gainsight does play a role here because results are surfaced from our search engine. But since we only fetch specific details from Zendesk (title, description, url, source) our search algorithm has less impact on federated content than it has on content in the community (to compare: for community content we have more metadata that is taken into account to surface the most relevant results, such as; likes, views, comments, recent activity etc.)

2 replies

Blastoise186
Forum|alt.badge.img
  • Helper ⭐️⭐️⭐️
  • February 10, 2025

Hi ​@mbuuck1 ,

Given that I don’t have an instance to play with myself, it’s a little tricky for me to say for sure but I did find something that might help a bit: https://support.zendesk.com/hc/en-us/articles/4408830243482-About-help-center-federated-search and I did manage to find a couple of communities that are using it, so I fiddled about with those.

It seems that Federated Search basically pings a request to the Zendesk API and lets ZD do most of the work, while the Search tool itself on this side basically just displays whatever comes back from Zendesk as-is. As far as I can tell, the behaviour seems to be controlled by ZD as the search results seemed to be identical to what ZD itself was returning using its native search.

Based on the Zendesk Docs, it certainly seems that way! So I don’t think there’s much Gainsight can do to customise/optimize things because it’s ultimately not in control of how external search tools work.


Forum|alt.badge.img+3
  • Author
  • Helper ⭐️⭐️
  • Answer
  • December 31, 2025

Updating my post with new information that I learned from ​@Yoeri on the behavior of federated search. He confirmed in this other post that:

The way how federated search works is that we use the API to fetch Zendesk content and add it to the search engine in Gainsight (so you have a ‘bucket’ of content both from Gainsight and Zendesk on which we apply the search). Gainsight does play a role here because results are surfaced from our search engine. But since we only fetch specific details from Zendesk (title, description, url, source) our search algorithm has less impact on federated content than it has on content in the community (to compare: for community content we have more metadata that is taken into account to surface the most relevant results, such as; likes, views, comments, recent activity etc.)