Explain the North Star Metric and propose one for a product
Tests your ability to connect user value to business outcomes. A great answer defines the NSM, proposes one for a product (e.g., Spotify), and justifies how it links customer value to business success. A red flag is picking a vanity metric like DAU or revenue.
How would you diagnose why a new feature isn't being adopted?
This tests your ability to diagnose a flat KPI. A great answer outlines a funnel (awareness, activation, usage) and combines quantitative data with qualitative insights from session replays. A red flag is proposing solutions without a diagnostic plan.
Translate 'increase engagement' into a technical measurement plan
This tests your ability to translate a vague business goal into a structured, measurable technical plan. Clarify the goal with the PM, define a primary metric and supporting metrics, then create an instrumentation spec.
How do you build a business case for technical debt?
This tests your ability to translate engineering problems into business impact. A strong answer quantifies the debt's cost (e.g., slower velocity), frames it as risk, and proposes a concrete payback plan like allocating 20% capacity.
Throughput vs. Velocity in Agile Planning
This tests your grasp of flow vs. estimation metrics. Define throughput as a count of delivered items and velocity as a sum of estimated points. Throughput measures actual output, making it better for forecasting. Red flag: claiming velocity is more accurate.
How do you coach a team to self-management?
Tests your grasp of situational leadership. A great answer uses a maturity model like Tuckman's stages to show how your coaching evolves from directive teaching (Forming) to strategic advising (Performing), and explains which techniques you retire.
Advocating to decentralize a deployment approval board
Tests your ability to influence change with data. A good answer frames deployments as frequent, time-critical decisions ideal for decentralization, proposes a phased rollout with metrics like cycle time, and defines new guardrails.
Manager Wants to Attend Your Sprint Retrospective. How Do You Respond?
Tests your grasp of psychological safety and stakeholder management. A good answer identifies the risk to open feedback, proposes other ways to inform the manager, and suggests a structured, one-time experiment if they must attend.
A story is too large for one sprint. What are your options?
Tests your grasp of incremental value delivery over just task completion. A great answer first re-validates priority with the PO, then discusses vertical splitting strategies and their trade-offs.
How do you handle a story that's too large for a sprint?
Tests your ability to apply agile principles pragmatically. A great answer prioritizes the Sprint Goal, collaborates with the PO to vertically slice the story into smaller valuable pieces, and then re-plans the sprint backlog.
Design an Architecture for Rapid Product Iteration
This tests your grasp of evolutionary architecture for uncertain markets. Outline how to support guided, incremental change across multiple dimensions (tech, data, security) using fitness functions to protect key characteristics.
How do you quantify the cost of technical debt?
This tests translating engineering problems into business impact. Calculate the ongoing time cost per sprint, estimate the fix cost, and present a breakeven point to frame the refactor as an investment.
How do you handle non-functional requirements in a product backlog?
This tests if you can make abstract quality goals concrete. A good answer covers making NFRs explicit backlog items, adding them to the Definition of Done, and creating technical stories. A red flag is treating NFRs as assumed work that doesn't need tracking.
Blocker vs. Impediment: How do you escalate an impediment?
This tests your proactivity in removing systemic friction. Define a blocker (full stop) vs. an impediment (drag). Explain how you surface impediments in retros, track them in a backlog, and escalate systemic issues to leadership.
How does management's role evolve in scaled agile frameworks like LeSS?
This tests your grasp of modern agile management, moving from command-and-control to enabling teams. Contrast Theory X (directing) with Theory Y (coaching). In LeSS, managers improve the system and teach, not assign tasks.
How would you create a probabilistic forecast for 40 stories?
This tests your ability to use statistical methods for forecasting. A great answer explains how to use historical throughput in a Monte Carlo simulation to generate a probability distribution of completion dates, not a single point estimate.
When Does Cycle Time Begin and End?
This tests your grasp of process metrics. Define Cycle Time as starting when active work begins ('In Progress') and ending when it's 'done' (code complete/merged), not when the ticket was created. A red flag is confusing this with customer-facing Lead Time.
Explain Little's Law and its application in Kanban
Tests your grasp of flow metrics beyond the formula. A great answer defines the law (Lead Time = WIP / Throughput), explains the trade-offs (e.g., more WIP increases lead time), and shows how to set WIP limits.
How would you use a spike to de-risk a story?
Tests your use of Agile spikes for de-risking, not for building features. A good answer defines the goal, sets a strict time-box, and clarifies the deliverable is knowledge (e.g., a POC), not production code. A red flag is merging spike code into main.
Explain backlog refinement: purpose, participants, and outcomes
Tests your grasp of continuous planning for predictability. A good answer defines the purpose (clarify, estimate, prioritize), participants (whole team + PO), and outcomes (a 'Ready' backlog). A red flag is describing it as a one-off pre-sprint meeting.