Amazon’s 1on1s are not casual check-ins — they are performance data engines. Most employees misuse them as status updates, but high performers treat them as real-time calibration tools. The difference isn’t frequency or agenda — it’s whether you’re using the 1on1 to shape narrative, not just share tasks.
1on1 at Amazon: How to Navigate Perf Review and Feedback
TL;DR
Amazon’s 1on1s are not casual check-ins — they are performance data engines. Most employees misuse them as status updates, but high performers treat them as real-time calibration tools. The difference isn’t frequency or agenda — it’s whether you’re using the 1on1 to shape narrative, not just share tasks.
Not sure what to bring up in your next 1:1? The Resume Starter Templates has 30+ high-signal questions organized by goal.
Who This Is For
You’re a new or mid-level Amazon employee (L4–L6) in a technical or product role, preparing for your next performance cycle. You’ve had 1on1s that felt one-sided, vague, or reactive. You want to influence your review outcome before it happens — not after. This isn’t for managers learning to lead — it’s for ICs learning to survive and scale within Amazon’s feedback-driven culture.
What is the real purpose of 1on1s at Amazon?
The real purpose of Amazon 1on1s is to create a written record of impact, not discuss blockers.
In a Q3 debrief for a L5 TPM candidate, the hiring committee approved promotion — but only after the candidate produced 18 months of 1on1 notes showing sustained ownership, escalation patterns, and peer mentorship. The bar wasn’t delivery — it was documentation. At Amazon, if it’s not written, it didn’t happen. Your 1on1 is not a conversation — it’s a legal trail.
Most employees treat 1on1s as weekly syncs: “Here’s what I did. Here’s what’s blocked.” That’s not insight — it’s inventory. The best employees use the same time to signal judgment: “I chose X over Y because of customer risk Z. I escalated A at day 3, not day 10. I coached two peers on roadmap trade-offs.” These aren’t updates — they’re evidence packages.
Not feedback, but audit readiness.
Not alignment, but narrative control.
Not rapport, but reputation engineering.
You’re not building trust — you’re building a case. Your manager isn’t your coach — they’re your auditor and advocate, often at odds with each other. The 1on1 is where you reconcile that duality by feeding them raw material for your advocate self.
> 📖 Related: Google PM Resume ATS vs Amazon PM Resume ATS: 5 Key Differences
How should you structure a 1on1 with your manager?
A structured 1on1 at Amazon must include four written sections: Impact, Escalations, Peer Influence, and Input Requests — in that order.
During a 2023 HC calibration for a L6 SDE promotion, one candidate was blocked because their 1on1 logs showed zero peer mentorship. Another was approved despite missed deadlines because their logs showed three cross-team escalations with data-backed rationale. The differentiator wasn’t output — it was dimensionality.
Amazon evaluates leadership via the LPs (Leadership Principles), and each 1on1 should reflect at least two per session. But don’t label them — demonstrate them. Writing “I showed Ownership” is worthless. Writing “I re-architected the deployment pipeline after owning postmortem for 3 SOC2 findings” shows it.
Your 1on1 doc should be templated, shared 24 hours in advance, and never exceed one page. Use this structure:
- Impact (1–2 bullets): Delivered outcome, not task. “Reduced EC2 overprovisioning by 37%” not “Ran cost audit.”
- Escalations (0–1 bullet): Show you know when to pull the cord. “Escalated latency SLA breach to L7 on Day 2 after internal comms failed” signals ownership.
- Peer Influence (1 bullet): Prove you lead without authority. “Coached L4 PM on DR roadmap during unplanned outage” hits Learn and Be Right.
- Input Requests (1–2 items): Force decision, not discussion. “Need your view on prioritizing Q4 tech debt vs. feature build — decision by Friday.”
Not “how’s it going,” but “here’s what I moved.”
Not “can I get help,” but “here’s where I need your voice.”
Not “what do you think,” but “here’s the decision I recommend.”
The doc isn’t for your manager — it’s for the HC member who’ll read your name for the first time in six months. Make it self-contained.
How do you use 1on1s to prepare for perf review cycles?
Perf reviews at Amazon are backward-looking, but 1on1s are forward-loading machines.
In a 2022 compensation review, two L5 product managers had identical project delivery. One received a 12% cash bonus, the other 7%. The delta? The first had 37 1on1 notes with consistent references to “customer obsession in trade-off decisions” — exactly the LP the comp band was calibrated against. The second had notes like “discussed roadmap blockers.” Same work, different narrative.
Amazon’s perf cycle runs on written artifacts. Your 1on1 history is the only source of truth for your manager when they write your justification. If your impact isn’t in the 1on1 logs, it doesn’t exist in cycle.
Start 6 months before review season. Tag each 1on1 with the LPs demonstrated. Use exact phrasing from the LP definitions — “insisted on high standards” not “held team accountable.” When your manager drafts your packet, they’ll copy-paste from your logs. Make it easy.
Not timing, but trail density.
Not visibility, but verbatim reuse.
Not memory, but material.
In one debrief, a hiring manager admitted: “I wrote her entire LoE from her 1on1 summaries. I didn’t remember the project — but the notes were clear.” That’s the goal. Become your manager’s source library.
> 📖 Related: Amazon RSU Vesting vs Google RSU Vesting: Which Is Better for Your Career?
How do you give upward feedback in Amazon 1on1s without risking your review?
Upward feedback at Amazon is high-risk theater — unless it’s framed as customer impact.
In a Q4 check-in, a L4 SDE told their manager: “You’re not giving me clear prioritization.” The manager rated them “meets expectations” in the next cycle. Same quarter, another L4 said: “Unclear Q3 priorities led to 3-week delay in fraud detection launch — $1.2M incremental loss per AWS’s model. Recommend bi-weekly priority lock with you.” That employee got promoted.
The content was identical. The framing made the difference.
Amazon protects upward feedback via the “Start, Stop, Continue” system in 1on1s. But raw use fails. “Stop: You’re not available” triggers defensiveness. “Start: A 15-minute priority sync every Monday would reduce downstream rework — like the 3-week drift in auth work” ties behavior to business cost.
Use data, not feeling. Cite customer metrics, not workload. Say: “Our CSAT dropped 18 points post-launch. I recommend we pause feature work to run a fix sprint — and I’d like your help unblocking support bandwidth.” Now it’s not criticism — it’s leadership.
Not “I need,” but “the customer needs.”
Not “you failed,” but “we missed a signal.”
Not “help me,” but “let’s fix it together.”
One L6 eng manager told me: “I ignore emotional feedback. But if someone ties a process gap to a business KPI, I act. Because I’ll get asked about it in my own review.” That’s the lever.
How often should you document and share 1on1s at Amazon?
You must document and share every 1on1 — no exceptions — and retain logs for 18 months.
Amazon’s performance system runs on retrieval. In a 2023 promotion appeal, a L5 PM was denied because their manager “couldn’t recall specific examples.” But the candidate produced 14 months of shared 1on1 notes. The HC overturned the decision — not because of the work, but because the paper trail existed.
Notes must be shared 24 hours before the meeting, stored in a manager-accessible folder, and never edited after the meeting. Real-time edits look like revisionism.
Frequency is non-negotiable: weekly for L4–L5, bi-weekly for L6. Missed 1on1s = missing data. One L6 candidate lost a promotion slot because they had only 8 1on1s logged in 6 months — manager cited “low engagement.” Reality? They’d met — but not documented.
Not trust, but traceability.
Not memory, but metadata.
Not intent, but indexability.
Your 1on1 log is your performance résumé. It’s scanned by HCs, comp teams, and skip-levels. Make it machine-readable: use consistent headers, avoid jargon, and include dates, project IDs, and LP tags.
Preparation Checklist
- Turn every 1on1 into a written doc with Impact, Escalations, Peer Influence, and Input Requests.
- Share the doc 24 hours in advance — never walk in cold.
- Tag each entry with 1–2 Leadership Principles using Amazon’s official phrasing.
- Store all 1on1 logs in a shared, dated folder — no edits post-meeting.
- Review logs quarterly to ensure at least 15 documented peer influences per year.
- Work through a structured preparation system (the PM Interview Playbook covers Amazon’s LP-driven feedback loops with real HC debrief examples).
- Schedule skip-levels every 90 days — bring one data-backed input request per meeting.
Mistakes to Avoid
BAD: “Here’s what I worked on this week.”
This is task reporting — not leadership. It provides zero signal for perf reviews. Managers see it as maintenance, not momentum. You’ll be labeled “reliable” — not “promotable.”
GOOD: “I re-prioritized the auth migration after discovering a P1 security gap — aligned with Be Proactive and Insist on Highest Standards.”
This shows judgment, LP alignment, and ownership. It’s reusable in review packets.
BAD: “Can you help me unblock X?”
This positions you as dependent. In HC discussions, managers describe such employees as “needing oversight.” That’s a career ceiling.
GOOD: “I’ve tried A and B. I recommend we escalate to L7 by EOD — here’s the customer impact if we delay.”
This shows rigor, escalation sense, and decision support. Managers repeat this verbatim in reviews.
BAD: Giving feedback: “You don’t give me enough autonomy.”
This sounds like a personal grievance. It triggers defensiveness and damages trust. You’ll be seen as high-maintenance.
GOOD: “Locking roadmap decisions bi-weekly reduced rework by 40% in Q2 — suggest we formalize the rhythm to scale it.”
This frames feedback as a tested insight with business impact. It’s not criticism — it’s scaling a win.
FAQ
Do Amazon managers read your 1on1 notes before perf cycles?
Yes — they pull directly from them to write your review. In a 2022 HC, a manager admitted they used 80% of a candidate’s 1on1 language in the LoE. If it’s not in your logs, it won’t be in your packet.
Should you bring prep docs to skip-level 1on1s?
Yes — bring one written input request with data. Skip-levels ignore vague asks. One L5 got promoted after showing a skip-level a 1on1 log proving repeated escalation delays. Data gets action.
Can poor 1on1s hurt your promotion chances?
Yes — even with strong delivery. In 2023, a L6 PM was delayed because their 1on1s showed no peer mentorship. The HC said: “No evidence of scale.” Your 1on1s must show dimension — not just depth.
Ready to build a real interview prep system?
Get the full PM Interview Prep System →
The book is also available on Amazon Kindle.