.jpg)
The measurement problem
Every technology company claims its engineers use AI. Few can prove it. 'We use GitHub Copilot' is a procurement statement, not a productivity measurement. The question is not whether AI tools are available. It is whether engineers are actually using them, how much of the output is AI-assisted, and whether the AI-assisted output meets quality standards.
How FourKites measures
Ticket-level attribution
Every AI-assisted coding session at FourKites is linked to a Jira ticket ID. This is not optional. The tooling requires a ticket ID to start a session. This creates a complete audit trail: for every ticket completed, FourKites knows whether AI was used, how much time the AI session consumed, and what output was produced. This attribution enables measurement at the individual, team, product line, and organization level.
Signature detection in PRs
AI-generated code carries identifiable signatures in commit metadata. FourKites tracks the percentage of PRs that contain AI-generated code, the percentage of lines that are AI-generated, and the review pass rate for AI-generated versus human-generated code. This data answers the critical quality question: is AI-generated code passing review at the same rate as human-generated code?
The weekly pulse
The CEO Engineering Pulse Report tracks AI adoption rate across 17 product lines. Each product line's AI adoption percentage is visible, creating transparent accountability. Teams with low adoption are not penalized; they receive support to identify barriers. Teams with high adoption are studied to identify practices that can be replicated.
Why this matters
Measurement transforms AI adoption from a cultural aspiration to an operational reality. Without measurement, 'AI-native engineering' is a marketing claim. With measurement, it is a quantified capability: what percentage of output is AI-assisted, what is the quality differential, and what is the throughput impact. FourKites can answer all three questions with data.
Newsletter
Stay Informed. Join 30,000+ monthly readers and get exclusive ebooks, reports, and industry insights from FourKites every week.
Thank you for your submission!
Oops! Something went wrong while submitting the form.
One billion hours of operational work completed by AI agents over the next decade.
The supply chains that adopt autonomous execution in the next 24 months will define the competitive standard for the next decade. The ones that do not will spend that decade trying to catch up.
Ready? Talk to an Outcome Advisor
