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

# Overview

> A monthly, prioritised action plan of GEO and paid media recommendations grounded in your real AI visibility data.

## The idea in one sentence

Opportunities gives you a monthly action plan — a prioritised list of GEO and paid media recommendations grounded in real AI visibility data.

Instead of digging through dashboards to figure out what to do next, you land on a clear list of actions that matter most, with the reasoning and execution briefs already prepared.

## Key principles

Opportunities are AI-generated recommended actions produced for your workspace, based on the **last 30 days of collected data**.

Core principles to keep in mind:

* **Monthly cycle.** When recurring generation is active, new opportunities are generated every month and made available on the **1st of each month**, always based on the last 30 days of data.
* **Existing recommendations are never erased.** New generations add opportunities; they don't delete previous ones.
* **Conditions must be met.** Each opportunity type has data requirements. If the workspace doesn't meet them, that type cannot be generated:

| **Opportunity type** | **Requirements (min.)**                     |
| :------------------- | :------------------------------------------ |
| Influence            | 1 active Search project                     |
| Organic & On-site    | 1 active Search project + GEO Audit project |
| Paid Media           | 1 active Brand Perception analysis          |

## What it covers

Recommendations span three areas, visible in the sidebar:

**Influence** : actions on platforms that shape what AI models say about your brand: YouTube, Reddit, and PR (news media, specialist blogs, comparison sites).

**Paid Media** : actions that use paid channels to fill AI visibility gaps: ChatGPT Ads, Amazon Rufus Ads, and DV360 programmatic.

**Organic** : actions on your own website to make your pages more citable by AI engines: page structure, readability, schema, images, and content gaps.

## Browsing opportunities: group by Team, Country or Brand

When you open Opportunities, you land on a **home page that groups your opportunities by dimension** so you can see where they are concentrated before diving into the detailed list.

A switcher lets you choose the grouping dimension:

* **Brand** — grouped by brand, each card showing the brand logo.
* **Country** — grouped by market, each card showing the country flag.
* **Team** — grouped by team, each card showing a team icon.

Each group card shows, at a glance:

* the **group name and icon** (brand logo, country flag, or team icon);
* the **total number of opportunities** in the group;
* a **status gauge** breaking the group down by status (To review / In my plan / Dismissed);
* the **average priority score** of the group — computed only over opportunities that are still *To review* or *In my plan* (dismissed ones are excluded);
* **related-dimension counts** that depend on the grouping you picked:
  * grouped by **Country** → number of Brands, Channels and Teams;
  * grouped by **Brand** → number of Countries, Channels and Teams;
  * grouped by **Team** → number of Countries, Brands and Channels.

Clicking a group card takes you to the detailed opportunities list, **pre-filtered on that group** — so a team lead can jump straight to their team, or a brand owner straight to their brand.

## How it works

Opportunities follow a simple flow, from decision to delivery:

1. [**Selecting opportunities**](/platform/opportunities/selecting-opportunities) — review each recommendation, add the relevant ones to your plan, and dismiss the rest.
2. [**Working on briefs**](/platform/opportunities/briefs) — once an opportunity is in your plan, its briefs are generated automatically; edit them with AI or manually, then copy or export them.
3. [**Tracking in Kanban**](/platform/opportunities/kanban) — follow each opportunity through its brief statuses on the Kanban board.

## What makes it reliable

Recommendations are driven by rules applied to real performance data — not creative guesses. The same gap in the data will always produce the same category of recommendation. What you see is reproducible and tied to a specific measurement window (the last 30 days of your Search project data).