Using SharePoint as primary knowledge base for JSM Help Center & Virtual Agent (without Confluence)?

[Tomislav Tobijas]

Hey everybody,

So, I have a question which related to SharePoint being the primary (and probably only) KB source for help center, portal (and potentially VA). I do understand that Help Center KB and VA are different things and areas, but I've decided to put them all together here.

Anyway, let's say we have this 👇

User story

As a client, when I have knowledge in SharePoint, I want to use it as the primary knowledge source so that I can self‑serve that knowledge in the JSM Help Center and via Virtual Agent.

Context

The client’s main knowledge base currently lives in SharePoint. They would prefer to keep SharePoint as the only KB and not migrate content to Confluence.

Acceptance criteria

• Knowledge in SharePoint should be accessible in the Help Center (objects are searchable, suggested as solutions when raising tickets, etc.).
• Customers should be able to use the Virtual Agent to access this knowledge.

Now, what I've 'found':

• Obviously, JSM's built‑in knowledge base is tightly integrated only with Confluence (for article search, suggestions, indexing, etc.)
• I could add external resources (incl. SharePoint URLs)
• The option would also be to use Help Center topics 

However, those solutions/links are not indexed like native KB articles.

There's also this feature request that partially covers what I'm looking for.

🤔 I was also thinking of using 'Confluence as a wrapper' - like use smart links to SharePoint content, or embedding SharePoint docs in Confluence pages, but I don't think these will be indexed as expected (still to test this part out).

❓ So, some questions here would be:

1. Has anyone successfully implemented a setup where SharePoint (*or any other source) remains the primary KB, but JSM customers can still effectively search and self‑serve via the Help Center and/or Virtual Agent?

2. Are you using any marketplace apps to:

   • index SharePoint content for JSM, or

   • make Virtual Agent respond based on SharePoint content?

3. Any best practices, pitfalls, or architectural patterns you’d recommend (e.g. partial sync to Confluence, middleware, custom search, etc.)?

4. If you tried to stay 100% on SharePoint and not use Confluence for KB at all, what worked and what didn’t?

Note that I'm also exploring/testing Customer Service Management to see what's possible and what's not there, primarily when it comes to AI-assistance. 💡

P.S. - We can either talk about Standard, Premium, or Enterprise plans. All paths are open.

Thanks,
Tobi

[Kris Klima _K15t_ · Accepted solution]

Hi @Tomislav Tobijas and welcome to the Community :D 

I wonder if you could engineer @Alexander Nilsson's solution in reverse.

https://community.atlassian.com/forums/Confluence-Cloud-Admins-articles/Automating-Confluence-PDF-Exports-for-Offline-Access-and/ba-p/3151528

Pulling individual content blocks from Sharepoint into Confluence pages programmatically and serving them in JSM.

[Tomislav Tobijas]

Hehe, thanks for the welcome @Kris Klima _K15t_ 😆

Totally forgot it might be simpler to ping you directly about any tips&tricks when it comes to this 😁

It's definitelly a cool solution. Nice one @Alexander Nilsson for sharing! 🙌

This is actually not a bad idea - we could do some kind of n8n automation that syncs stuff from SharePoint to Confluence. The only problem might be formatting, but I guess it's doable

Thanks for the idea, Kris! 

[Kris Klima _K15t_]

 @Tomislav Tobijas 

@Matt Reiner _K15t_ and me are doing a live stream ACE class next week about migrating things into Confluence, including pre/post migration checks, it may be helpful (for your clients too).

[Matt Reiner _K15t_]

Joinnnnn Usssss!