AICPB Rankings Methodology

 

AICPB provides a standardized, data-driven framework for ranking global AI products and AI models based on usage, performance, and operational metrics. This page outlines the methodology used to collect data, define metrics, and produce AICPB rankings across different platforms and categories.

 

1. Purpose of the Rankings

The goal of AICPB rankings is to offer an objective, comparable view of AI adoption and performance worldwide.

AICPB rankings focus on measurable signals rather than subjective evaluations or self-reported claims.

 

They are designed for:

• Researchers and analysts

• Media and industry observers

• Investors and founders

• Developers and AI practitioners

AICPB publishes two distinct but complementary ranking systems:

• AI Product Rankings — measuring real-world user

• AI Model Rankings — measuring model-level performance and operational characteristics

 

2. Data Collection

AICPB aggregates data from multiple sources, including:

 

• Own Technology-based Data Collection and Analysis

• Mobile application analytics providers

• Web traffic measurement platforms

• Publicly available disclosures

• Verified third-party industry datasets

 

All data undergoes normalization and validation processes to ensure consistency across regions and platforms.

 

3. AI Product Ranking Metrics

3.1 Monthly Active Users (MAU) — Mobile Applications

For mobile applications, rankings are primarily based on Monthly Active Users (MAU).

 

Definition:

MAU represents the number of unique users who actively used a mobile application within a given calendar month.

 

Scope:

• Includes iOS and Android apps combined

• Excludes all website traffic and web-based usage

MAU is used to rank products in App-based rankings only.

 

3.2 Website Monthly Visits — Web Applications

For web-based products, rankings are based on Website Monthly Visits.

Definition:

Website Monthly Visits represent the total number of user visits to a product’s website within a given month.

Scope:

• Includes desktop and mobile web traffic

• Excludes mobile application usage

Website Monthly Visits are used to rank products in Web-based rankings only.

 

4. AI Product Platform Separation

AICPB intentionally separates rankings by platform:

• Mobile Applications: Ranked by MAU

• Web Applications: Ranked by Monthly Visits

Metrics are never combined across platforms. This ensures that each ranking reflects comparable user behavior.

 

5. AI Model Rankings

In addition to product-level rankings, AICPB publishes AI Model Rankings focused on model performance and operational characteristics.

AI model rankings evaluate how models perform in real-world deployment environments rather than abstract theoretical benchmarks alone.

 

5.1 Model Ranking Dimensions

AI models are ranked across one or more of the following dimensions:

• Price
  Cost efficiency based on publicly available pricing information and standardized usage assumptions.

• Latency
  Time taken to generate responses under typical operational conditions.

• Throughput
  Number of requests or tokens processed per unit of time.

• Uptime and Availability
  Reliability and operational stability over time, based on observed service availability and incident data.

Each dimension is ranked independently to avoid conflating different performance characteristics.

 

5.2 Model Scope

AI Model Rankings apply to:

• Publicly available commercial AI models

• Models accessible via APIs or hosted inference services

Model rankings do not measure:

• Model capabilities via subjective scoring

• Training data quality or proprietary internals

• Vendor marketing claims

 

5.3 Update Frequency for AI Models

AI Model Rankings are updated on a weekly basis, ensuring timely and comparable insights into performance and operational changes.

Each model ranking page clearly indicates:

• The ranking dimension

• The relevant update period

 

 

6. Categories and Classification

Each AI product may appear in one or more functional categories, such as:

• AI Chatbot

• AI Search

• AI Productivity Tools

• AI Image Generation

• AI Coding Assistants

Category inclusion is determined based on a product’s primary use cases and user-facing functionality. More Product Categories → Visit:

👉 AI Rankings

 

7. Geographic Scope

Unless otherwise specified, AICPB rankings are:

• Global

• Aggregated across all available regions

Regional rankings may be introduced separately in the future.

 

8. Update Frequency Summary

• AI Product Rankings → Updated monthly

• AI Model Rankings → Updated weekly

Each ranking page explicitly states the applicable time period to avoid ambiguity.

 

9. Limitations and Notes

• Rankings represent relative position, not absolute market size.

• Data availability may vary by region and platform.

 

10. Data Usage and Citations

All ranking results published by AICPB are governed by the AICPB Data License, which defines permitted uses, redistribution terms, and attribution requirements.

Researchers, journalists, and analysts using AICPB rankings should follow the official AICPB Data Citation Guidelines to ensure proper attribution and version accuracy.

• Data License: https://www.aicpb.com/data-license

• Citation Guidelines: https://www.aicpb.com/citation-guidelines