CFD 'Code Review' band is widening. What does it mean?

What's really being asked

This question tests your ability to move from data to diagnosis to action. It's not just about reading a chart. It's about demonstrating systems thinking—understanding that a symptom (a widening CFD band) points to a deeper process issue. Interviewers want to see if you can use metrics to identify a bottleneck, hypothesize its root causes, and propose specific, non-judgmental, and measurable interventions.

The full answer

A strong answer has three parts. First, the diagnosis: clearly state that a widening band means the arrival rate of work into 'Code Review' is greater than the departure rate. This increases the Work In Progress (WIP) for that stage and, consequently, the average time a ticket spends in that stage. Second, the investigation: before jumping to solutions, list potential causes to investigate. This could include PRs being too large, an insufficient number of available reviewers, low-quality code requiring multiple review cycles, or slow CI feedback delaying the process. Third, the action plan: propose specific, measurable actions tied to the potential causes. For example, suggest an experiment to limit PR size, schedule dedicated 'review focus blocks' for the team, or improve pre-review automated checks to catch issues earlier.

The mistakes people make

A major red flag is jumping directly to a single, expensive solution like "We need to hire more senior engineers." Another is blaming people, such as saying "Developers need to be less picky" or "They just need to review faster." This shows a lack of process-oriented thinking. Vague answers like "We need to improve our review process" are also weak because they lack specific, actionable steps. Finally, simply stating that the band is widening without explaining the underlying mechanics of arrival vs. departure rates shows a surface-level understanding.

What usually comes next

Expect follow-ups like: "How would you validate which of those potential causes is the real one?" (Answer: by looking at other data like PR size distribution, time to first comment, or number of review cycles per PR). Or, "You propose limiting PR size. What would you expect to see on the CFD in a month?" (Answer: The 'Code Review' band should stabilize or narrow, and overall throughput should increase). A third might be, "What if the team resists these changes?" (Answer: Frame it as a time-boxed experiment, present the data clearly, and facilitate a discussion to gain buy-in).

A concrete example

A widening band means lead time for that stage is growing. If 10 tickets enter 'Code Review' per week but only 7 leave, the WIP for that stage grows by 3 tickets weekly. After a month, you have 12 extra tickets queued, and the average review time has ballooned. My first step would be to check our PR metrics. If the average PR size grew from 200 to 800 lines of code over the last quarter, that's a likely culprit. I'd propose a two-week experiment to keep all PRs under a 250-line threshold and measure the impact on the CFD's 'Code Review' band.