All Collections
Fin AI
Fin AI Copilot
Prepare your workspace for Fin AI Copilot
Prepare your workspace for Fin AI Copilot

How to optimize your knowledge base for AI and set Fin and your teammates up for success.

Beth-Ann Sher avatar
Written by Beth-Ann Sher
Updated over a week ago

Fin AI Copilot uses machine learning to source answers from your support content, including past conversations, public articles, internal articles, and snippets. Before using Fin with your support team, it’s worth taking some practical steps to ensure the knowledge is available and optimized for AI.

Fin will work with our new Knowledge feature for managing all content. You'll automatically be given access to this feature when your workspace is added to the Fin AI Copilot beta. Join the waitlist if you haven't opted in yet.

Below are some steps you can take to set Fin and your teammates up for success while you wait for access to the Fin AI Copilot beta.

Step 1: Ask content owners to review their content

If you’ve got existing support content in your knowledge base, the first thing you should do is review and ‘cleanse’ that content. AI is only as good as the knowledge you give it, so if your content is out of date, the answers Fin provides your teammates will be less accurate too.

Check for accuracy and update or archive content where necessary. From sales and engineering, to legal and security - knowledge can be created by multiple teams and subject matter experts (SMEs). So divide your content by owner and ask each team or SME to review their knowledge area.


  • We find it helpful to export all content into a spreadsheet where you can easily organize it by team, and all teams can collaborate off the same sheet by marking off each item they’ve reviewed. This gives you a clear overview of the current state of content and the progress made.

  • Reviewing all support content at once can be a daunting task! Focus on the most used content first i.e. the content with the most views. Employ the 80/20 rule here and ensure that your more popular content is up to date first.

Step 2: Reformat and template your content

Once all your content has been fact-checked by experts, you should start looking at how each article is structured. When creating content for AI, there are some formatting best practices that can make a big difference between making the answer available, and allowing Fin to easily find and use that answer in the right context:

  • Use headers

  • Use rich formatting (tables, bulleted lists, etc.)

  • Avoid ambiguity

  • Restate questions

  • Give context to multimedia

Create a style guide and reusable templates for different types of content, then share them with knowledge managers to ensure all new content follows this best practice for AI.

Step 3: Identify gaps in your content

Ensure you’re not missing any important content. Use reporting to find “Searches with no results” and see what customers or teammates have been unable to find in your current knowledge base. You should also talk to your support team and other stakeholders and get them to inform on knowledge gaps or which resources they’d like to have more of.

Step 4: Build an Information Architecture

Fin AI Copilot can access both internal and public articles from the Knowledge hub. With all knowledge coming together under one roof, you need to build a solid Information Architecture (IA).

You'll be given access to the new Knowledge hub once your workspace has been added to the Fin AI Copilot beta.

A good IA organizes content in a clear and logical way to enable teammates to navigate and find content quickly, and helps maintain consistency when new content is added because every piece of information has a clear home.

This means creating a hierarchical folder structure which could look something like this:

If you have multiple knowledge managers or teams who own content, we recommend using Optimal Workshop where your teammates can sort cards into categories that make sense to them. This can help you uncover an intuitive structure that works well for everyone.

If your existing knowledge base lives in Guru, Notion or Confluence, good news - you’ll be able to import it all across to your new Knowledge hub and keep your existing IA 😃

Step 5: Import/sync your content with the Knowledge hub

The new Knowledge hub will automatically be available to all customers on the Fin AI Copilot beta. Once you have access to the beta, you can start importing or creating new content directly from here.

You can also decide which of this content you want to allow teammates to use with Fin AI Copilot.

You must have the Fin AI Copilot beta enabled on your workspace before you can access the Knowledge hub. Join the waitlist to get access.

Bonus: Create a feedback loop for continuous improvement

After investing in a tip-top knowledge base to power Fin AI Copilot and your support team, you don’t want to let content become stale and unhelpful. This will not only negatively impact Fin’s performance, but also reduce all teammate efficiency gains you made by having this knowledge readily available and accessible.

Consider implementing a feedback loop which enables your support team (or even your wider org) to flag inaccuracies, suggest improvements, and request new content to fill any knowledge gaps.

We use Back-office tickets to enable our teammates to submit these requests straight from the Help Desk. If Fin gave an answer that didn't help, they can check the source(s) used and open a ticket to get the content improved without leaving the inbox or losing too much time.

Our knowledge team ensures these requests are actioned within a week and lets the teammate know by responding to the ticket and closing the loop. This way, teammates are helping themselves and each other find accurate answers faster by giving Fin improved content to use in the future.


Need more help? Get support from our Community Forum
Find answers and get help from Intercom Support and Community Experts

Did this answer your question?