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

# Generative Engine Optimization (GEO) & AI Search Intelligence

> How YPYM DBX tracks brand citations across Google AI Overviews, ChatGPT Search, and Perplexity with geographic precision and strict cost governance.

## Executive Summary & Market Context

The organic search paradigm has fundamentally transitioned from traditional link lists to AI-synthesized responses. Modern decision-makers interact directly with **Google AI Overviews (Gemini in SERP)**, **ChatGPT Search**, and **Perplexity AI** to evaluate enterprise vendors, financial solutions, and business services.

In this environment, ranking on page one of Google is no longer sufficient if generative AI engines synthesize answers using competitor citations while completely omitting your brand.

**YPYM DBX (Digital Brand Experience)** delivers an enterprise-grade **Generative Engine Optimization (GEO) Suite** that monitors brand presence, citation health, and generative share of voice across major AI platforms without compromising financial governance or data integrity.

```mermaid theme={null}
flowchart TD
  A["Target Search Query"] --> B["Geographic Localization Engine\n(Criterion ID: 2360 / Indonesia Proxy)"]
  B --> C["Google SERP + AI Overview\n(Daily Batch - $0 Extra Cost)"]
  B --> D["Deep GEO LLM Scraper\n(On-Demand with Approval)"]
  C --> E["Citation Extractor & Domain Matcher"]
  D --> E
  E --> F["Brand Share of Voice & Visibility Scoring"]
  F --> G["YPYM DBX Minimalist Executive Dashboard"]
  
  style A fill:#1A4BFF,stroke:#1A4BFF,color:#fff
  style B fill:#30FFFC,stroke:#1A4BFF,color:#000
  style C fill:#1A4BFF,stroke:#1A4BFF,color:#fff
  style D fill:#f8f9fa,stroke:#e9ecef,color:#333
  style E fill:#1A4BFF,stroke:#1A4BFF,color:#fff
  style F fill:#30FFFC,stroke:#1A4BFF,color:#000
  style G fill:#1A4BFF,stroke:#1A4BFF,color:#fff
```

YPYM DBX Generative Engine Optimization (GEO) multi-tier telemetry tracking Google AI Overviews, ChatGPT Search, and Perplexity with verified Indonesian geographic origin.

***

## 1. Supported AI Search Platforms & Extraction Architecture

| Platform | Extraction Method | Telemetry Points Captured | Cost Profile |
| :- | :- | :- | :- |
| **Google AI Overviews (Gemini)** | Native SERP API parser (`ai_overview`) | Synthesis text, cited URLs, source domains, title tags, and position relative to organic | **Included (\$0 extra)** via daily crawl |
| **ChatGPT Search** | Live UI Scraper (`chat_gpt/llm_scraper`) | Formatted markdown, citation cards, external hyperlinks, product tables, sponsored ads | On-Demand (\~Rp 19 - Rp 25) |
| **Perplexity AI** | Live LLM Web Grounding (`perplexity/llm_responses`) | Real-time web citations, source domain authority, reasoning synthesis | On-Demand (\~Rp 58 - Rp 70) |
| **Google Gemini Standalone** | Enterprise Flash Model (`gemini/llm_responses`) | Entity recognition, topical context, brand sentiment | On-Demand (\~Rp 18 - Rp 25) |

***

## 2. Indonesian Geographic Precision & Anti-Bias Grounding

A fundamental challenge in AI search tracking is **geographical hallucination**. When a prompt regarding financial or regulatory services is dispatched without strict geographical parameters, international LLM endpoints default to North American or European entities.

YPYM DBX guarantees geographic data integrity through three-layer localization:

1. **Google SERP & AI Overview**: Bound to Google Ads Criterion ID `2360` (Indonesia) or metropolitan city codes (e.g. `1007716` Jakarta, `1007710` Tangerang) with `hl=id` & `gl=ID`.
2. **ChatGPT LLM Scraper**: Enforces `location_code: 2360` and `language_code: "id"` routed through verified Indonesian residential proxies.
3. **Multi-LLM Responses**: Leverages `web_search_country_iso_code: "ID"` and `web_search_city: "Jakarta"` to force search grounding within Indonesian regional domains (`.id`, `.co.id`, `.go.id`).

***

## 3. Two-Tier Cost Governance Model

To safeguard enterprise budgets against runaway token consumption, YPYM DBX separates tracking into coordinated operational tiers:

```
                      ┌──────────────────────────────────────────────┐
                      │             Keyword Tracking Model           │
                      └──────────────────────┬───────────────────────┘
                                             │
                    ┌────────────────────────┴────────────────────────┐
                    ▼                                                 ▼
      [ Tier A: Standard (Included) ]                   [ Tier B: Multi-LLM GEO Intelligence ]
     Google SERP + Google AI Overviews                 ChatGPT Search, Gemini, Perplexity, Claude
     • Daily automated crawl at 11:00 AM WIB           • Daily automated batch crawl at 11:30 AM WIB
     • Zero extra API expenditure                      • Top action bar "Crawl all Prompts" (On-Demand)
     • Continuous daily organic baseline               • 6-digit cryptographic OTP confirmation required
```

***

## 4. Brand Visibility & Share of Voice (SOV) Ranking Algorithm

YPYM DBX evaluates commercial dominance through a position-weighted point formula:

$\text{Score}(\text{Position}) = \max(0, 11 - \text{Position})$

* **Top Organic Positions**: Rank 1 earns 10 pts, decreasing linearly to Rank 10 (1 pt).
* **Google AI Overview Citation**: $+15 \text{ Poin}$ premium for top-of-page generative placement.
* **ChatGPT Primary Citation**: $+12 \text{ Poin}$ premium for conversational recommendation.

This quantitative score enables multi-tenant portfolio managers to track domain-level **Share of Voice (SOV)** and benchmark competitor emergence over time without manual competitor registry.

***

## 5. Dual-Tab Executive Interface Architecture

To maintain serene luxury minimalism and avoid analytical clutter, YPYM DBX bifurcates search intelligence into two dedicated tabs:

### 5.1. Tab 1: SERP + AIO (`/domain/[slug]`)

* **Primary Scope**: Continuous daily ranking telemetry for Google Organic SERP alongside embedded Google AI Overviews.
* **Top Action Bar**: Co-located action controls featuring **Crawl History**, **Crawl SERP Keywords** (protected by 6-digit OTP verification at \$0.0035/kw), and **+ Add Keyword**.
* **SERP Top 10 Penetration Trajectory**: 100% Stacked Bar visualizer mapping the proportion of keywords ranking in Google Top 10 against unranked or sub-page-1 positions.
* **Google AI Overview (AIO) Citations Heatmap (Dark Mode)**: 90-day (3-month) dynamic rolling matrix rendered in Deep Obsidian Dark Mode. Features automated daily date advancement (+1 day rolling forward with oldest day retirement), 5-tier citation intensity coloring, day-of-week indexing (Mon–Sun), and rich hover inspection with cited prompt listings.
* **Clean Position Typography**: Positions beyond Google Top 100 or non-ranking queries are rendered as a minimal dash (`-`), maintaining pristine visual cleanliness without noisy `>100` badges.
* **Reciprocal Brand Matching**: Fully symmetric host matching guaranteeing consistent citation verification across multi-level brand assets (such as subdomains and apex domains).
* **Drawer Inspection**: Interactive side panel detailing full reference citations and Gemini source URLs.

### 5.2. Tab 2: LLM Citations & GEO (`/domain/llm/[slug]`)

* **Primary Scope**: Cross-engine conversational intelligence covering **ChatGPT Search**, **Google Gemini Standalone**, **Perplexity AI (Sonar)**, and **Claude AI**.
* **Clean State by Default**: Zero mock data by default; displays real database records alongside a schedule status indicator (*"Daily automated crawl at 11:30 AM WIB"*).
* **Unified Top Action Bar**: Co-located action controls featuring **Crawl History** modal inspection, **Crawl all Prompts** (auditing 4 AI engines at \~\$0.0089/prompt protected by 6-digit OTP verification), and **+ Add Prompt**.
* **Crawl Execution History**: Centralized log drawer accessible from the action bar displaying recent SERP and Multi-LLM crawl runs, keyword counts, duration, and engine states.
* **Executive Preview Mode**: Discreet switcher toggling between **Live Data** and **Demo Preview (21 Days)** for executive presentations and stakeholder reviews.
* **Right-Aligned Header Controls**: Mode toggles (**Trend**, **Comparison**, **Stacked**) and engine filter pills (**All Engines**, **ChatGPT**, **Gemini**, **Perplexity**, **Claude AI**) are consolidated side-by-side on the right side of the card header, eliminating vertical whitespace bloat.
* **Prompt-First Terminology**: Unified English phrasing across the interface—referencing queries as **Prompts** (`All Prompts`, `Filter Prompts...`, `Prompt / AI Query`) rather than generic keywords.
* **Dual-Metric Telemetry**: Prominent citation counts paired with overall brand citation share across monitored prompts.
* **Multi-Mode Trajectory Visualization**:
  * **Trend (Line)**: 1.5px smooth spline curves displaying trajectory velocity and day-over-day citation progression.
  * **Comparison (Grouped Bar)**: Side-by-side grouped bars for direct head-to-head comparison across the four engines on each date.
  * **Stacked (Stacked Bar)**: Stacked bars highlighting cumulative brand citation volume and total engine share.
* **Spacious Canvas Geometry**: High-resolution visual canvas delivering crystal-clear Y-axis integer ticks, trajectory curves, and non-overlapping tooltips.
* **Engine-Specific Visual Hierarchy**: Distinct, brand-native color coding across OpenAI (ChatGPT Search), Google (Gemini Standalone), Perplexity AI, and Anthropic (Claude AI) for effortless multi-platform comparison, supported by clean monochromatic badges maintaining serene luxury minimalism.
* **On-Demand Audit Modal**: Interactive per-keyword audit modal providing real-time multi-engine citation checks with instant source URL inspection and transparent pre-flight cost verification.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.