Effective collaboration and high-quality code reviews are essential to maintaining a strong engineering culture.

This guide provides a structured approach to monitor team interactions, evaluate review dynamics, and spot potential risks in knowledge distribution and code review standards.

1️⃣ Quantity

Analyze Reviewer-Submitter Relationships

➡️ Where: PR Insights → Collaboration Tab

What to look for:

• Use the submitters-reviewers graph to visualize how individuals interact through code reviews.
• Ideally, you'll see connections from every reviewer to every submitter (though this is rare).
• Check for isolated patterns, such as one submitter consistently reviewed by only one person—this may indicate a knowledge silo or a 'bus factor' risk.

📘

The number of reviews may not match the number of PRs. One PR can have multiple reviewers or none.

Evaluate Review Coverage and Knowledge Sharing

➡️ Where: PR Insights → Collaboration Tab → Sharing Index & Number of Reviews Graphs

Challenges to consider:

• An increase in reviews might mean more PRs OR more reviews per PR.
• KSI typically increases with more reviews but can drop if PRs surge and reviews lag behind.
• In general, KSI and total number of PRs are inversely proportional, because when you open more PRs, it's hard to keep KSI up.

🚧

Sudden drops in KSI might reflect intense delivery periods where collaboration took a backseat.

2️⃣ Quality

Monitor Code Review Quality

➡️ Where: Merge Quality
Merge quality metrics offer a snapshot of how thorough and effective reviews are.

Suggestions:

• Watch for trends that suggest declining attention to review standards.

Key Takeaways

• Strong collaboration is reflected in broad, balanced reviewer-submitters graphs.
• Watch for uneven review patterns — they may signal risks to knowledge continuity.
• Interpret KSI in the context of overall activity; drops could indicate a focus on speed over collaboration.
• Merge quality can highlight where review standards might need reinforcing or refining.