AI Analytics - analyze your data with AI, safely and correctly

WitCloud MCP connects your All In One data with AI models like ChatGPT, Claude, or Gemini, enabling natural language data analysis.


📘 Introduction – What is “AI Analytics”?


AI Analytics is a module that allows you to analyze your All In One (AIO) data using large language models such as ChatGPT, Claude, or Gemini.

Instead of writing SQL queries or building dashboards, you can ask business questions in natural language and get clear, explainable answers - based on your existing analytical data model.


Unlike generic AI tools, AI Analytics is designed specifically for structured business data.
It works exclusively on prepared AIO analytical tables, follows strict usage rules, and respects your business logic.


The goal is simple:
let AI analyze data - without breaking your data model or business logic.


🎯 Module goal – why did we create “AI Analytics”?


Many companies already have clean, unified data in All In One - but accessing insights still requires:

  • SQL knowledge
  • dashboard maintenance
  • time-consuming ad-hoc analysis

At the same time, generic AI tools:

  • do not understand your data model
  • guess table meanings
  • ignore business-specific rules
  • and often produce misleading results

That’s why we created AI Analytics.


AI Analytics combines:

  • the reliability of the All In One data model
  • with the flexibility of AI-driven analysis

This allows teams to explore data, answer questions, and validate hypotheses faster - without sacrificing correctness or control.


⚙️ How does it work?


AI Analytics uses WitCloud MCP (Model Context Protocol) - an internal control layer that defines how AI interacts with your data.

The process can be described in a few simple steps:

  1. Understanding the data model
    AI reads the All In One introduction to understand how your data is structured.
  2. Applying business context
    If available, project-specific business rules and definitions are loaded.
  3. Selecting the right tables
    Only predefined AIO analytical tables are used - raw data is never queried.
  4. Reading documentation first
    Before any analysis, table documentation is read to ensure correct interpretation.
  5. Safe analysis and explanation
    Queries are generated according to strict rules and results are explained in plain language.

You don’t need to manage this process - it happens automatically.


🧠 Core concepts


All In One as a single source of truth

AI Analytics works exclusively on data prepared in the All In One (AIO) module.

AIO:

  • unifies data from advertising, analytics, and e-commerce systems
  • standardizes metrics and dimensions
  • provides ready-to-use analytical tables

AI never works on raw data or unknown structures.


WitCloud MCP

WitCloud MCP controls how AI accesses your data.

Its role is to:

  • guide AI to the correct AIO tables
  • enforce documentation and usage rules
  • apply business-specific context
  • prevent incorrect or unsafe analysis

You can think of MCP as a guardrail system for AI-powered analytics.


Documentation-first approach

Before AI analyzes any table, it always reads its documentation.

This ensures that:

  • metrics are interpreted correctly
  • dimensions are used as intended
  • analyses remain consistent with the AIO data model

This approach replaces guesswork with clarity.


💵 Pricing

AI Analytics is available as an optional module and is billed based on actual usage.

The fee is charged monthly only for users who executed at least one query via WitCloud MCP during the given billing period.


💰 Price per user (tiered model)

  • 1 user → $30 / month
  • Users 2–5 → $25 / user / month
  • Users 6+ → $15 / user / month

Pricing follows a progressive (tiered) model — the more active users in a given month, the lower the unit price applied to higher tiers.



📌 Pricing examples


✅ Example 1

In a given month:

  • 1 user executed queries via MCP

➡️ Total active users: 1
💵 Cost breakdown:

  • 1 × $30 = $30

👉 Monthly AI Analytics cost: $30



✅ Example 2

In a given month:

  • 4 users used AI Analytics

➡️ Total active users: 4
💵 Cost breakdown:

  • User 1 → 1 × $30 = $30
  • Users 2–4 → 3 × $25 = $75

👉 Monthly AI Analytics cost: $105



✅ Example 3

In a given month:

  • 8 users executed queries via MCP

➡️ Total active users: 8
💵 Cost breakdown:

  • User 1 → 1 × $30 = $30
  • Users 2–5 → 4 × $25 = $100
  • Users 6–8 → 3 × $15 = $45

👉 Monthly AI Analytics cost: $175