> ## Documentation Index
> Fetch the complete documentation index at: https://docs.alvahq.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Knowledge Base

> AI-extracted sales knowledge from your email conversations.

The Knowledge Base is where Alva stores insights extracted from your email conversations — objection handling techniques, product feature discussions, pricing patterns, competitor intelligence, and more. Over time, it becomes a library of what works in your sales process.

## Accessing the Knowledge Base

Go to **Knowledge Base** in the sidebar.

## How knowledge is extracted

When Alva [analyses your emails](/email/understanding-email-analysis), it doesn't just look for sentiment and signals — it also extracts **reusable knowledge**. For example:

* A prospect objected to your pricing, and your rep successfully reframed the value → that's an **objection handling** entry
* A prospect asked about a specific use case, and your rep explained it well → that's a **use case** entry
* A competitor was mentioned in a deal you won → that's **competitor intel**

Alva extracts these automatically and adds them to your Knowledge Base as draft entries.

## Categories

Knowledge entries are organised into six categories:

| Category             | What it covers                                           |
| -------------------- | -------------------------------------------------------- |
| **Product Features** | How your team describes and positions features           |
| **Objections**       | Common objections and how they were handled              |
| **Use Cases**        | Specific use cases and how they were explained           |
| **Pricing**          | Pricing discussions, discounting patterns, value framing |
| **Competitor Intel** | Competitor mentions, comparisons, win/loss context       |
| **Other**            | Everything else that doesn't fit the above               |

## The review workflow

Knowledge entries go through a review process:

<Steps>
  <Step title="Draft">
    Alva extracts an insight and saves it as a draft. Drafts need human review before they're considered reliable.
  </Step>

  <Step title="Approved">
    A team member reviews the draft and approves it. Approved entries are trusted knowledge that Alva can reference in future conversations.
  </Step>

  <Step title="Archived">
    Entries that aren't useful or are outdated can be archived. They're removed from active use but kept for reference.
  </Step>
</Steps>

## Reviewing entries

From the Knowledge Base page, use the tabs to switch between **Draft**, **Approved**, and **Archived** entries. For each entry, you can:

* **Approve** — mark it as reliable knowledge
* **Archive** — remove it from active use
* **View the source** — see which deal and email thread it came from

## Scoring

Each entry has two scores:

* **Confidence score** — how confident Alva is that this is a real insight (not noise)
* **Effectiveness score** — how successful the approach was (based on deal outcome and email responses)

Entries are sorted by these scores by default, so the most reliable and effective knowledge appears first.

## How Alva uses approved knowledge

Approved knowledge entries feed back into Alva's responses. When you ask Alva for help with an objection, a use case, or competitive positioning, it draws on your Knowledge Base to give advice that's specific to your business — not generic AI-generated suggestions.

<Tip>
  Review your Knowledge Base drafts regularly — especially after closing deals. The insights extracted from won deals are the most valuable, as they capture what actually worked.
</Tip>