Methodology

How CrowdWiseAI scores and ranks AI tools.

CrowdWiseAI measures adoption, developer activity, and ecosystem growth across leading AI tools using publicly observable signals.

Scoring v0.7c

Methodology Status

Scoring Version

v0.7c

Data Points Tracked

30+

Scoring Signals

18

Data Sources

10

Tools Tracked

944

Update Cadence

Daily

Scoring Overview

Each tool receives a CrowdWiseAI Score (0–100) computed from 18 scoring signals across 4 pillars. Signals are normalized to remove outlier effects, then combined into a weighted composite score adjusted for data coverage.

The CrowdWiseAI Momentum Score measures the rate and direction of change in a tool's signals over time. Tools are classified as Surging, Rising, Stable, or Falling based on their momentum trajectory.

Pillar Weights

Adoption 35%
Maintenance 30%
Friction 20%
Ecosystem 15%

4 Pillars × 18 Signals

Adoption

35% of final score
GitHub StarsGitHub ForksPyPI Downloadsnpm DownloadsStack Overflow ActivityGoogle Trends InterestWeb Traffic Rank

Maintenance

30% of final score
Commit RecencyIssue HealthRelease Cadence

Friction

20% of final score
Developer Friction ScoreActivity Recency

Ecosystem

15% of final score
GitHub ForksGitHub StarsDownstream DependentsContributor CountStack Overflow PresenceHacker News Mentions

Data Coverage

Not all signals apply to every tool. For example, a Python-only library has no npm downloads — it is not penalized for this. CrowdWiseAI computes coverage as the ratio of available signals to applicable signals for each tool, and applies a bounded adjustment so that missing data has a measurable but limited impact on the final score.

Tools with low data coverage are flagged and excluded from momentum-based rankings to prevent noisy signals from affecting the market narrative.

Tool Universe

CrowdWiseAI tracks tools across two tiers:

  • Infrastructure — Frameworks, libraries, and platform-level software
  • Applications — End-user tools, agents, RAG, evaluation, and workflows

Tools are organized into 22 categories. Each category shows its average score, tool count, and aggregate momentum.

Data Sources

CrowdWiseAI collects data daily from public developer ecosystems and web platforms. All signals are derived from publicly accessible sources.

GitHub
PyPI
npm
Stack Overflow
Google Trends
Hacker News
Reddit
YouTube
Web Traffic Rankings
Hugging Face

Data is timestamped and archived daily. Each ingestion produces a provenance receipt documenting the source, fetch time, and data snapshot.

Tool names, logos, and trademarks belong to their respective owners.
CrowdWiseAI is an independent research and analytics platform.

Questions about the CrowdWiseAI methodology or data sources can be directed to contact@crowdwiseai.com.