Metrics question

One of Decagon's largest customers has launched an agent, but adoption has plateaued because internal teams will not let it handle higher-value interactions. How would you diagnose whether the bottleneck is model quality, workflow design, integration gaps, or change management, and how would you decide which intervention to make first?

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

Start a mock interview on this question · Mock interview from a job description

What this question tests

Tests diagnosing an adoption plateau caused by human trust limits rather than a technical failure, and choosing the right intervention.

How to approach it

  1. Separate the four candidate causes explicitly: model quality (accuracy on high-value tasks), workflow design, integration gaps, and change management.
  2. Check model quality first with task-level accuracy data on the higher-value interactions specifically, not aggregate accuracy.
  3. If accuracy is already strong, interview the internal teams directly about why they still route manually, since this points to trust or change management.
  4. Check workflow design: does the agent even have a clear handoff path for higher-value cases, or is escalation the only option shown to agents.
  5. If model and workflow are sound, treat this as a change-management problem, and design a graduated trust rollout, starting with human-reviewed agent suggestions.
  6. Prioritize the intervention with the clearest evidence rather than defaulting to a model fix, which is the easiest default.

What a strong answer includes

Common mistakes

Likely follow-up questions

More metrics questions

More questions from Decagon

Learn the skill behind it

Chapters of the AI PM course that teach what this question tests.

Preparing for a specific role?

Book summaries for this kind of question

Browse all 4,000+ questions in the bank