A knowledge base turns the answers your team already has into something users can find on their own. It powers self-service, cuts down repetitive tickets, and gives everyone a single, reliable source of truth.
This guide walks through how to create a knowledge base in InvGate Service Management: building and publishing articles, generating and improving content with AI, importing existing articles, and the practices that keep the whole thing useful over time. If you are evaluating knowledge base software, it doubles as a hands-on look at how the pieces fit together.
How to create a knowledge base in InvGate Service Management
Creating a knowledge base comes down to organizing your content into categories and publishing articles that reach the right people. Here is the flow from start to publish.
1. Set up your categories
Start by creating the categories that will hold your content. Go to Settings > Knowledge Base > Categories and create a category in the root or nest it inside another to build subcategories, so the structure mirrors how your users actually look for answers.
When you create a category, you can link it to your Service Catalog. This is where a category earns its keep: with featured articles, you can attach knowledge base articles to a catalog category, and those articles surface automatically the moment a user starts a request in that category. Users get a proven answer before they ever open a ticket, which is one of the most direct ways a knowledge base reduces request volume.
2. Create an article
Go to Articles and click New Article. You choose the category, set who owns the article, give it a title and an optional description that doubles as search keywords, and write the content in the editor. The editor supports images, videos, audio, file attachments, links, and more, so a single article can cover anything from a quick FAQ to a full troubleshooting walkthrough.
Set the article's visibility to match its audience:
- Private access: agents, coordinators, and administrators only.
- Registered users: everyone above, plus end-users.
- Public access: generates a shareable link so people who are not registered in the system can view the article.
This lets you keep internal runbooks restricted to agents and open self-service guides to everyone, all from the same knowledge base.
3. Publish and let it work
Once published, the article is available in your knowledge base, and InvGate Service Management proactively suggests it whenever a user creates a request on a related topic.
From the published view, users can rate the article, follow it to get update notifications, save it as a favorite, and download it as a PDF. Your team gets usage signals back, including views, ratings, and how often an article resolved a request, so you always know what is working.
Using AI to create and improve knowledge articles
Writing clear, complete articles by hand takes time that competes with everything else on your plate. InvGate Service Management puts AI directly inside the knowledge base to speed up writing, capture solutions automatically, and keep content ready for self-service.
Improve your writing without leaving the editor
AI-powered content enhancement lets you summarize, expand, refine, or adjust the tone of any selected text right in the article and snippet editor. You do not switch to another tool to polish a draft or tighten a long explanation, and you stay in control the whole time: AI-generated changes are not applied until you confirm them, and nothing goes live until you save or publish it yourself. You can use these improvements when writing a new article or editing an existing one.
Turn resolved tickets into articles
When a ticket is resolved, the option to create a knowledge article appears at the top of the screen. Click Generate, pick the relevant messages and resolution steps, and the AI drafts an article from the request and the work that solved it in under 30 seconds. Your team reviews, refines, and publishes. This turns everyday support work into reusable content instead of knowledge that disappears when the ticket closes. It is part of AI Hub, which you enable from Settings > AI Hub.

Grow your knowledge automatically with Knowledge Discovery
Knowledge Discovery analyzes closed tickets on a recurring basis and turns them into structured knowledge fragments called Snippets, so your knowledge layer grows even when no one sits down to write an article. It follows a human-in-the-loop model: administrators review, edit, accept, or reject each Snippet, and visibility is organized in levels:
- Not visible: Snippets stay in the administration section for moderators only. They are not approved and do not power any AI feature.
- Agent users: Snippets appear to agents as part of Solution Recommendations.
- Virtual Service Agent: Snippets reach end-users through the Virtual Service Agent, along with agent Solution Recommendations.
Conversational self-service
Your published articles and your approved Snippets both feed the Virtual Service Agent, which resolves requests conversationally across the self-service portal, Microsoft Teams, WhatsApp, and Slack, and both surface to agents as Solution Recommendations. When AI Hub is enabled, an AI Availability field on each article shows whether the latest version has been indexed and is in use, so you always know your edits are live where it counts.
How to import knowledge base articles into InvGate Service Management
If your knowledge base is just getting started, or you already have valuable documentation in other tools, you do not have to copy it over by hand. You can import knowledge base articles into InvGate Service Management, and it preserves the original formatting and creates the articles for you.
In the Articles section, the + button at the top left opens three options:
- Create article: opens a blank editor.
- Import articles: bring in one or more articles from .md, .txt, .doc, .docx, .pdf, or .html files.
- Import articles via CSV: load HTML content for multiple articles at once through a single CSV file.
The options are available from Settings > Knowledge Base > General, with the Import articles button next to Save. Upload a single file, several files together, or one CSV covering many articles, and your existing content lands in the knowledge base ready to organize and publish. For large, programmatic migrations, a public API is also available to create and manage articles in bulk.
Knowledge base best practices in InvGate Service Management
A few habits keep your knowledge base fast to navigate and genuinely useful:
- Design categories around how users search, and link them to the Service Catalog so featured articles surface at the moment a request is created.
- Set visibility deliberately. Use private, registered, and public levels so each audience sees exactly the content meant for them.
- Capture knowledge where the work happens. Generate articles from resolved tickets and let Knowledge Discovery surface Snippets from closed tickets, so documentation keeps pace with real operations.
- Bring in what you already have. Import existing documentation rather than rewriting it, then refine each article with AI content enhancement in the editor.
- Keep content indexed for AI. After editing an article or Snippet, check the AI Availability field so Solution Recommendations and the Virtual Service Agent serve the latest version.
- Write for scanning. Keep articles concise, lead with the solution, and add visuals and links to related knowledge base articles so users can move through a topic in sequence.
Ready to build yours?
Creating a knowledge base in InvGate Service Management is a matter of organizing your categories, publishing articles with the right visibility, and letting AI write, capture, and reuse the solutions your team produces every day. Do it well, and your knowledge base becomes the backbone of self-service and of every AI feature that depends on it.
Want to try it hands-on? Sign up for a 30-day free trial.