# Custom Knowledge Training

Feed your company's existing knowledge into your AI employees so they answer questions and make decisions using your real data, not generic guesses. Train from a connected app (Notion, Slack, HubSpot, and 65+ more), a website crawl, an uploaded file, or pasted text. Training is company-wide: every employee shares the same knowledge base and searches it automatically on every conversation.

Your AI employee only becomes valuable when it knows your business. Upload company documents, product specs, brand guidelines, process manuals, and internal policies. It reads, indexes, and understands everything, then uses that knowledge every time it works. The result is an employee that answers like someone who has been at your company for years.

Training sources go beyond simple file uploads. Connect Notion workspaces, Google Docs folders, Confluence spaces, and other knowledge bases. The training pipeline extracts content, processes it into structured knowledge, and indexes it for instant retrieval. When your employee writes a blog post, it pulls from your actual brand voice doc. When it answers a customer question, it references your real product documentation.

Connection is not an all-or-nothing import. When a source exposes pages, folders, channels, projects, or records, select exactly what enters training. If a source cannot expose a useful list, Sistava asks for explicit confirmation before importing all accessible content.

Training is incremental. Add new documents at any time and your employee absorbs them without losing what it already knows. When a product spec changes, retrain the selected source or schedule an automatic refresh. Your knowledge base stays current without starting from scratch.

Works with PDFs, Word docs, text files, web links, and connected SaaS tools. No special formatting required. Feed it what you have, and it figures out the rest.

## Train Your AI Employees on Your Actual Business Knowledge

A generic AI agent knows what its training data covers. A Sistava employee knows your products, your processes, your customers, and your brand, because you trained it on your own documents. Upload PDFs, Word files, internal wikis, and SOPs, and the agent ingests them as working knowledge.

Knowledge training is not keyword search. The AI employee builds a semantic understanding of your content, so it can answer nuanced questions, make connections across documents, and apply your knowledge to novel situations it has never seen before.

## Connect Live Sources: Notion, Google Docs, and More

Static file uploads are useful, but your knowledge is always changing. Sistava connects directly to Notion workspaces and Google Docs so you can retrain selected source content without manual re-uploads. You can also schedule automatic refreshes for a completed training run.

This connection matters most for fast-moving information: pricing sheets, product specs, policy documents, and process guides. Retrain when information changes so your AI employee works from the current version you selected.

You can combine multiple knowledge sources for a single AI employee. A Customer Success agent might pull from your Notion knowledge base, your product changelog in Google Docs, and a set of uploaded case studies, all active simultaneously.

## Choose What Enters Company Knowledge

A connected account may be able to read a broad set of information, but training does not have to use all of it. Choose the pages, folders, channels, projects, or records that belong in your company's shared knowledge before a training run begins.

OAuth controls what the provider permits the connection to access. Import selection is the next control: it decides which accessible items are actually processed into knowledge. If an integration cannot provide a meaningful item list, you must explicitly confirm an all-accessible-content import.

## Knowledge That Stays Inside Your Team

Training builds one company-wide knowledge base shared by your AI employees. Choose the right source material before import so the shared base reflects the information your team should use.

This source-selection control is separate from per-employee tool access. You can enable a connected app only for the employees that need to act in it, while the training choice controls what the company-wide knowledge base includes.

## How Do I Train an AI on My Company's Documents?

Upload the files or connect the source, and the pipeline does the rest: content is extracted, processed into structured knowledge, and indexed for retrieval. There is no labeling, no fine-tuning project, and no data science involved. If you can attach a file, you can train an employee.

Training is also never finished in the best sense: drop in next quarter's pricing or a new process doc whenever it exists, and the employee folds it into what it already knows. The knowledge base grows the way a good hire's understanding does, continuously and without ceremony.

## Grounded Answers Instead of Confident Guesses

The biggest fear with AI in front of customers is the confident wrong answer. Training changes the failure mode: the employee answers from your documents, cites what it actually knows, and stops guessing about the things you never told it.

That grounding is what makes delegation safe. A support reply quoting your real refund policy and a proposal using your actual pricing are not lucky outputs. They are retrievals from the knowledge you supplied.

## How It Works

**Upload documents or connect Notion and Google Docs to give your AI agent deep knowledge of your specific context.**

Training works by ingesting your documents into a dedicated knowledge graph for the agent. You can upload PDFs, paste text, or connect your Notion workspace and Google Docs directly. The agent builds a semantic index over everything you share, so it can retrieve the right piece of knowledge in any conversation, not just keyword-match against raw text.

Trained knowledge persists across all conversations and channels. Whether the agent is responding in chat, handling an email, or executing a scheduled task, it draws from the same knowledge base every time. You can add new documents at any time and the agent incorporates them immediately, no retraining or restarting required.

## Use Cases

### Customer success team training an agent on product docs

Upload your documentation and the AI employee answers questions with accurate, company-specific knowledge from day one.

### Legal team training an agent on internal policies

Load contracts, policies, and procedures. The AI agent references the right document before taking any action.

### Sales team giving an agent competitive and product context

Train the AI employee on pitch decks, battle cards, and product sheets. Every response reflects your current positioning.

### Support team keeping an agent current as products evolve

Retrain selected materials or schedule an automatic refresh so the AI employee can use updated product knowledge.

## Comparison

| Before | After |
|---|---|
| Generic AI tools give generic answers that don't reflect your business. | Train the AI employee on your docs and it answers with your knowledge. |
| Keeping an agent updated requires technical intervention. | Retrain selected materials or schedule a refresh when your knowledge changes. |
| Agents hallucinate when they don't know the answer. | Grounded in your knowledge base, the agent stays accurate and on-topic. |
| New employees take weeks to learn the product. | Your AI agent arrives pre-trained and ready to use your knowledge from day one. |

## FAQ

### Can I choose which connected content is imported?

Yes. For sources that expose pages, folders, channels, projects, or records, select exactly what enters training. If an integration cannot provide a useful selection list, Sistava asks you to explicitly confirm importing all accessible content.

### What file formats can I upload for knowledge training?

Sistava supports PDF, DOCX, TXT, and Markdown files. You can also connect Notion databases and Google Docs directly as live knowledge sources. Support for additional formats is expanding.

### How long does it take for an AI employee to learn from uploaded documents?

Processing typically completes within seconds to a few minutes depending on document size. Once processing is complete, the AI employee can immediately draw on that knowledge in conversations and tasks.

### Does selecting content limit trained knowledge to one AI employee?

No. Training is company-wide, so the items you select become shared knowledge for your AI employees. Use source selection to control what enters that shared knowledge base.

### What happens when I update a connected Notion page or Google Doc?

Retrain the selected source when its documents change, or schedule an automatic refresh after a completed training run. You do not need to re-upload the documents manually.

### Can my AI agent learn from my company's existing documents and knowledge base?

Yes. You can upload docs, connect Notion, Google Docs, and other sources, and the AI employee uses that knowledge when answering questions and completing tasks. Training updates take effect immediately.

### Do I need to clean up or format documents before uploading?

No. PDFs, Word docs, text files, and links work as they are. The pipeline extracts and structures the content for you.

### Does trained knowledge apply on every channel?

Yes. The same knowledge base serves chat, email, Slack, voice, and scheduled tasks, so the employee gives consistent answers everywhere.

### Can several employees train on the same documents?

Yes. Shared knowledge lets one set of documents serve the whole team, without re-uploading per employee.

> I uploaded our 80-page product spec and connected our Notion wiki. Now the agent answers questions about our product better than most people on the team.
>
> Marcus L., Product Lead · SaaS company

## Where Custom Knowledge Training fits

Custom Knowledge Training is part of Things that make them smart.

Upload documents, connect Notion or Google Docs, and choose the source content that becomes shared company knowledge. A 6-layer memory system ensures they never forget a conversation, a decision, or a preference. When one employee learns something new, the whole team benefits through shared knowledge.

- [Things that make them smart](/en/features/knowledge): Learns everything. Forgets nothing.

## Read the guide

- [Guide: Custom Knowledge Training](/en/guide/equip/train)

## More in Knowledge

- [Remember Every Conversation](/en/features/knowledge/conversational_memory): Your AI employees remember what happened in past conversations, across your whole workspace, not just the one they are in right now. Preferences, decisions, and facts shared in chat get carried forward automatically, so you never have to repeat yourself to the same employee, or explain the backstory to a different one. It works out of the box, with nothing to set up or turn on.
- [Keep Notes Between Tasks](/en/features/knowledge/persistent_notes): Every employee keeps a personal notebook of what it has learned about you: your preferences, corrections, contacts, and how you like work done. That notebook loads automatically at the start of every conversation, so you never have to repeat yourself. Update a preference once and it sticks for good, across every future task and every channel.
- [Train from File Uploads](/en/features/knowledge/training_from_file): Upload a PDF, Word doc, Excel sheet, PowerPoint deck, plain text, Markdown, CSV, JSON, or XML file straight into Training, and every employee on your team can search it within minutes. Up to 20 files per run, each up to 72 MB, feed the same shared knowledge base your employees check on every message. No connected app and no crawlable URL required.
- [Academy Courses](/en/features/knowledge/academy_courses): Sistava Academy is a set of free, self-paced courses that teach you how to get real work out of an AI employee. Each course is a short series of written lessons you read in the browser, no video required, followed by a quiz. Pass the quiz and you get a certificate of completion with your name on it, ready to download as a PDF.
- [Shared Team Knowledge](/en/features/knowledge/shared_team_knowledge): What one AI employee learns in conversation, every employee in your workspace can draw on, automatically. Tell your sales lead about a new pricing rule and your support employee already has it the next time a customer asks, with no retraining and no copy-pasting the same context into each employee. Train an employee on a set of documents (a policy, a playbook, a product spec) and that knowledge is searched before every response across the workspace, so new hires start from what the team already knows instead of a blank slate.
- [Train From Any Source](/en/features/knowledge/knowledge_sources): Train your AI employees on the documents and tools your company already uses: upload files, point at any URL, or connect an app like Notion, Google Drive, Slack, or Zendesk. Training happens in place, with no exporting and re-uploading, and it teaches the whole workspace at once, not just one employee.

## Explore

- [Every feature](/en/features)
- [Hire an AI employee](/en/market)
- [Pricing](/en/pricing)