AI Active Users Metric

Definition

AI Active Users measures the number of unique developers who performed at least one AI-related interaction during the selected time range.

A developer is considered “AI active” if they triggered at least one qualifying AI interaction event within the selected period.

Qualifying interactions depend on the selected AI dimension (e.g., Coding Assistance, AI Coding, Code Review, Coding Agents).

What Counts as an AI Interaction

An AI interaction may include:

• Generating or inserting AI-suggested code
• Accepting AI-generated edits
• Using AI-assisted review suggestions
• Executing AI-driven coding agent actions

The exact qualifying actions depend on integration configuration and attribution logic.

How the Metric Is Calculated

AI Active Users is calculated as:

Count of distinct developers with ≥1 qualifying AI interaction event within the selected time range

Each developer is counted once per time range, regardless of how many interactions they performed.

Normalization

The headline value is normalized to the selected time bucket.

For example:

• Daily view → Unique AI users per day
• Weekly view → Unique AI users per week
• Monthly view → Unique AI users per month

The headline value represents:

Average number of unique AI-active developers per selected time bucket

This enables comparison across time ranges.

How the Metric Is Displayed in the Dashboard

1. Headline Value (e.g., 25 on average per week)

Represents the average number of unique AI-active developers per selected time bucket.

This is not a cumulative total.

2. Time-Based Values in the Chart

Each chart point represents:

The number of distinct developers who performed at least one AI interaction within that specific time bucket

Users are deduplicated within each bucket.

What This Metric Reflects

AI Active Users reflects:

• Adoption breadth
• Tool penetration
• AI usage participation
• Engagement distribution across teams

It does not measure:

• Volume of AI interactions (see Total AI Actions)
• Code impact (see Total Lines Added / Accepted)
• Productivity impact
• AI effectiveness

Relationship to Other AI Metrics

AI Active Users measures participation, not output.

Data Sources

Derived from:

• AI tool interaction logs
• Attribution signals
• Git metadata correlations
• Integration-level AI event tracking

Limitations

• Measures participation, not impact.
• A single minimal interaction qualifies a user as active.
• Attribution depends on AI detection accuracy.
• Shared accounts may distort counts.
• Small teams may show volatility.

AI Active Users reflects adoption coverage, not performance improvement.

Stakeholder Use Cases

Engineering Managers

• Monitor AI adoption across teams.
• Identify uneven AI usage patterns.

Team Leads

• Detect low adoption within specific squads.
• Track adoption over time.

Platform / DevEx Teams

• Measure AI rollout effectiveness.

Product / Executive Leadership

• Assess AI adoption breadth at organizational level.