BQ — Leadership Principles Standard Set

Every Amazon round runs 20-30 minutes of LP-anchored BQ. The recurring stems across 2026 loops are tight deadline, conflict, dive deep, out-of-scope contribution, tough feedback (given and received), disagree-and-commit, calculated risk, and customer obsession. Requirements Cover all 16 LPs with at least one STAR story each; have backup stories for the high-frequency LPs. Be ready for follow-up depth: "what would you do differently," "dive deeper," "what was the trade-off," "who pushed back." Tie every story to a quantified business outcome. Examples Verbatim stems pulled from 2026 Q2 loops: "Describe a time when you had a tight deadline but multiple workstreams." "Conflict with a peer; how did you resolve it?" "Out-of-responsibility work you took on." "Tough feedback you received / gave." "Tell me about a time you handled an ambiguous requirement." "Disagree and commit — when did you push back, then deliver?" "Risky decision you made under uncertainty." "Quick decision under ambiguity." "Dive deep on a project — bring numbers." "Describe how you organized a project or product end-to-end." "Tell me about a time you missed a deadline." "A time you held your position when you believed everyone else was wrong." "What technical optimization did you deliver for a customer?" "Describe a technical problem you solved for a customer." "Tell me about a time you turned a seemingly impossible project into something achievable." "Describe a time you learned something new very quickly." "Tell me about a time you refused a customer request." "Tell me about a time you improved a product by better understanding user needs." "Describe a time you delivered under a tight deadline, including the sacrifice, quality impact, and final outcome." Manager-loop prompts: "Tell me about a time you dealt with a low performer." "How do you balance challenging work and team health?" "How do you motivate people working on the team?" Notes HM / Bar Raiser rounds drill a single story 5+ layers deep. Pick stories where you genuinely owned the outcome and remember exact numbers. Several interviewers reward proactive clarification ("I'm glad you ask that"); treat under-specified BQs the same way you treat under-specified coding prompts. FAR / Premium-tier teams probe relentlessly on "what would you do differently" — every story needs a credible iteration beat. Each interviewer is assigned 1-3 LPs in advance and scores you on whether you meet / raise / fall below the bar for those specific principles — the same story can be reused across rounds because the lens differs. Customer Obsession is the de-facto north-star LP; weave a customer outcome into every story even when the prompt targets a different LP. Keep roughly 80% of the narrative in first-person 'I' actions. Stories that drift into 'we' / 'the team' lose the individual-contribution signal Bar Raisers look for. The Bar Raiser and the hiring manager are the only interviewers with veto authority — they own the LP and behavioral signal more than the technical signal. For MLE loops, the project deep-dive and LP evaluation may explicitly judge whether prior work had enough impact and scale. Prepare the numbers behind scope, users, revenue, latency, or model-quality movement, not just the technical narrative. Current VO loops keep LP in every technical slot, including coding and system design. Prepare for a short LP opener or closer even when the calendar invite names the round as purely technical. In-person NG loops now run four back-to-back rounds in roughly four hours and can pack 15+ LP prompts across them — most rounds interleave ~30 minutes technical with ~30 minutes BQ, and the final round is pure BQ. A single loop can sweep ten or more LPs (Have Backbone; Disagree and Commit, Invent and Simplify, Insist on the Highest Standards, Are Right A Lot, Deliver Results, Dive Deep, Customer Obsession, Ownership, Learn and Be Curious, Earn Trust), so story-bank breadth matters as much as depth — and plan physically: candidates describe losing their voice by round four. Preparation Build a 16x2 STAR matrix: 16 LPs × 2 stories each, all with metric outcomes and a 'do differently' answer. Drill the 'dive deep' follow-up by recording yourself answering the same BQ for 5 minutes, then again for 10. Pre-rehearse 'why Amazon' and 'why this team' — these still show up surprisingly often. Source candidate stories by listing the 3-5 most impactful projects from the last few years, then map each to 2-3 LPs it can serve — most candidates over-prepare unique stories per LP when one strong story flexes across several. Drill the metric layer: for every story, memorize the exact baseline number, the delta, and the business impact. "Data-driven" is the explicit rubric word in Amazon's own prep guidance. Practice the 'circular narrative': open with the problem, spend the body on personal actions with metrics, close by returning to how the problem was resolved. This holds up better under 10-15 minutes of dive-deep probing than a linear STAR recitation.