TL;DR

A 30-60-90 day plan for a new Amazon manager isn’t about proving competence — it’s about demonstrating judgment within the company’s 16 Leadership Principles. The best plans show deliberate trade-offs, not comprehensive task lists. Most fail by focusing on execution instead of influence, missing that Amazon promotes based on long-term leadership signaling, not short-term wins.

Who This Is For

You are a newly promoted or externally hired manager at Amazon, likely at the L5 or L6 level, earning between $182,000 and $250,000 total compensation. You report into a senior leader who expects you to “dive deep” within your first 30 days but also deliver clear “bar raiser” judgment by day 60. Your success isn’t measured by hitting targets — it’s measured by whether your presence changes team behavior and decision quality. If you’re still asking what to do, you’re already behind.

What Should a New Manager Focus on in the First 30 Days at Amazon?

In the first 30 days, your job is not to fix anything. Your job is to calibrate to the team’s existing operating rhythm and identify where the real friction lives. I’ve sat in HC meetings where a new manager was praised not for launching a process, but for explicitly choosing not to — because they recognized the team was already overloaded. That’s the signal Amazon wants: judgment over initiative.

Too many new managers come in with pre-built 30-60-90 templates downloaded from the internet and start checking boxes. In a Q3 org review, one L6 PM presented a polished plan with 22 initiatives. The VP paused and said, “Which three things are you saying no to?” The PM didn’t have an answer. He was not given a promotion cycle that year.

The first counter-intuitive truth is this: Amazon values omission more than inclusion.

Why? Because the Leadership Principle “Earn Trust” starts with listening, not acting. A new manager who rushes to optimize is signaling impatience, not leadership. The real work in month one is decoding the unwritten rules — who really makes decisions, where the data breaks down, and what past failures still haunt the team.

Not credibility through output — but credibility through diagnosis.

Not “what I’ll do” — but “what I’ve learned and why it surprises me.”

Not speed — but discipline in narrowing scope.

Your plan should be anchored in qualitative insights from 1:1s, not metrics from your first readout. In one debrief, a hiring manager said, “She didn’t even look at the dashboard in her first two weeks. But she surfaced a pattern in customer complaints that engineering had ignored for months. That’s bar raiser behavior.” That manager was backed for L7 within six months.

Your deliverable isn’t a project plan — it’s a stakeholder map with clear power dynamics annotated. Who resists change? Who enables it? What past initiative failed and why? That becomes the foundation of your real 30-day output: a diagnostic memo identifying one high-leverage constraint.

How Do You Align with Amazon’s Leadership Principles in Your 60-Day Plan?

By day 60, you must shift from observer to shaper — but only around one Leadership Principle that is clearly under-indexed in your team. Most managers try to demonstrate all 16 at once. That’s a failure pattern we explicitly screen for in HC reviews. When a candidate lists “Deliver Results” and “Invent and Simplify” and “Ownership” as areas of impact, we ask: which one are you willing to sacrifice for the others?

The second counter-intuitive truth: Alignment isn’t about checking boxes — it’s about making trade-offs visible.

Amazon doesn’t reward balanced leadership. It rewards asymmetric leadership — someone who will go to war for one principle, even if it creates short-term pain elsewhere.

In a hiring committee debate last year, an L5 manager pushed back on a roadmap change that would have improved speed-to-market but degraded long-term scalability. He framed it as “Thinking Long Term over Deliver Results” — and cited specific tech debt that had burned the team two years prior. The committee promoted him, not because he won the argument, but because he named the trade-off.

Your 60-day plan must contain at least one documented decision where you chose one Leadership Principle over another. Not “we did both” — but “we chose X, and here’s the cost.” That’s what “Bias for Action” really means — not moving fast, but moving decisively, with accountability.

Too many managers write “I collaborated across teams” in their self-review. That’s baseline. The signal is: “I escalated a conflict that others avoided.” One L6 wrote in her 60-day update: “Disagreed and committed with the finance lead on roadmap timing. Documented concerns in the PRFAQ appendix.” That line alone triggered a skip-level offer.

The third counter-intuitive truth: Amazon rewards documented dissent, not consensus.

Your 60-day output should include a PRFAQ, a small experiment, or a post-mortem that surfaces a systemic flaw. Not a success story — a learning story with teeth.

Use the 60-day mark to test a hypothesis: “If we change X, will decision velocity improve?” Then measure it. One new manager ran a six-week experiment shifting stand-ups from 15 minutes daily to 30 minutes biweekly with pre-reads. He proved a 22% reduction in meeting fatigue and a 17% increase in action-item completion. That wasn’t about meetings — it was a proof point for “Invent and Simplify.”

What Metrics Should You Track in Your First 90 Days as an Amazon Manager?

The right metrics aren’t the ones on your team’s dashboard — they’re the ones that reveal decision quality. Amazon leadership doesn’t care if you improved throughput by 15% if it came at the cost of long-term agility. In a recent HC review, we rejected a candidate who had “shipped 4 major features” because none had been stress-tested against customer escalation trends.

You must track leading indicators of leadership, not lagging indicators of output.

Not feature velocity — but rate of autonomous decision-making by ICs.

Not bug count — but time to root-cause resolution.

Not NPS — but percentage of negative feedback that triggers a process change.

The fourth counter-intuitive truth: Operational metrics lie — behavioral metrics don’t.

Your core dashboard should include:

  • % of ICs presenting at weekly tech reviews (measure of psychological safety)
  • # of escalated decisions per week (measure of team confidence)
  • Cycle time from problem identification to experiment launch (measure of bias for action)
  • Number of PRFAQs with “Disagree and Commit” annotations (measure of healthy conflict)

In one org, a new manager tracked “skip-level meeting follow-ups” — how many actions were taken after he relayed frontline feedback to his boss. In 90 days, he drove that from 20% to 68%. That wasn’t a project — it was a leadership proof point. He was flagged as high potential.

Also track your own influence decay. In month one, ICs will tell you what they think you want to hear. By day 90, they should be challenging your assumptions. One manager used a simple scoring system: he rated every 1:1 on whether the IC brought new information or just confirmed his view. When the ratio flipped from 3:7 to 7:3 in his favor, he knew he had shifted from evaluator to trusted partner.

Not trust through accessibility — but trust through impact.

Not “I held 30 1:1s” — but “5 engineers changed their roadmap priorities based on our conversations.”

And never present metrics without a judgment call. One L6’s 90-day review included: “Conversion increased 12%, but we believe this is a short-term anomaly due to seasonal traffic. We recommend holding off on scaling until Q2.” That restraint — backed by data — was what triggered his promotion nomination.

How Do You Build Credibility with Your Team and Stakeholders at Amazon?

Credibility at Amazon isn’t earned through likability — it’s earned through precision in escalation. I’ve seen HC discussions where a manager with low “team sentiment” scores was still promoted because he had “escalated one critical issue that prevented a P1 outage.”

You build credibility not by solving everything — but by knowing what deserves escalation.

Not “I fixed X” — but “I brought X to the right person at the right time with the right context.”

In a 2023 debrief, a new manager documented a 3-email thread to his VP: first, a data snippet; second, a one-paragraph implication summary; third, a proposed decision window. No attachments, no meetings — just a clear ask. The VP later said, “That was the most effective escalation I’ve seen in two years.” The manager was given a broader scope within six weeks.

Your first 90 days should include at least three deliberate escalation exercises — not crises, but tested judgment calls. Example: “Noticed a gap in customer verification logic. Flagged to security team with low urgency. Tracked resolution time and documented process lag.” That’s not about the bug — it’s about showing you understand risk tiering.

Also, invest in reverse onboarding. Most managers focus on understanding their boss. The differentiator is understanding their boss’s boss. One L5 sent a one-pager to her director within 14 days: “Here’s how I interpret your top three priorities, and where I see misalignment in our current roadmap.” The director replied, “You’re the first manager in two years to ask this.” That relationship became her sponsorship pipeline.

Not alignment through compliance — but alignment through challenge.

Not “I’ll support your goals” — but “Here’s where I think we’re off-track, and here’s a test to prove it.”

And never underestimate the power of the handwritten note. In a post-mortem, one manager sent individual thank-you cards to every IC who stayed late during a launch — not for the work, but for the “ownership mindset.” That detail came up in his HC review as proof of “recognizing exceptional performance.”

How Do You Demonstrate Leadership Beyond Your Role in the First 90 Days?

Leadership beyond your role isn’t about volunteering for extra work — it’s about creating reusable judgment. Amazon promotes people who generate leverage for others, not just for their own team.

The fifth counter-intuitive truth: Your first 90 days should produce one artifact others adopt.

Not a status report — but a template, framework, or escalation protocol that spreads organically.

In 2022, an L6 created a “PRFAQ Lite” format for small experiments — a half-page version of the standard document. It was adopted by three adjacent teams within six weeks. That wasn’t process for process’s sake — it was reducing cognitive load. He was promoted to L7 on that single artifact.

Your contribution should solve a cross-cutting pain point: meeting fatigue, PRFAQ bloat, standup inefficiency. One manager built a “decision log” — a shared doc tracking key calls, rationale, and owners. It reduced repeat debates by 40% in one team. Senior leaders started asking for copies.

Not visibility through presentation — but visibility through adoption.

Not “I presented at all-hands” — but “my template is now used in three orgs.”

Also, sponsor a junior IC in a high-visibility setting. Not by doing the work for them — but by giving them the platform. One new manager handed over a monthly metrics review to a junior BA, then sat in the back and took notes. The BA presented cleanly. The manager’s skip-level said, “That’s how you develop people.” That moment was cited in the promotion packet.

And finally, write something. Not a blog post — but a memo, a post-mortem, a principle statement. Amazon runs on written communication. If you haven’t written something that outlives your meeting, you haven’t led.

Preparation Checklist

  • Schedule 1:1s with every IC and peer within first 10 days; take notes in narrative format, not bullet points
  • Identify the team’s most recent failure and interview three people about why it happened
  • Draft a 1-pager on your boss’s top three priorities and validate it with them
  • Run one small experiment (e.g., meeting format change, doc template) and measure behavioral impact
  • Write a PRFAQ for a hypothetical project — even if it’s not approved — to test your bar level
  • Work through a structured preparation system (the PM Interview Playbook covers Amazon leadership mechanics with real HC debate examples)
  • Document at least one “disagree and commit” moment with rationale and follow-up

Mistakes to Avoid

BAD: "I optimized the sprint planning process and reduced meeting time by 30%."

This focuses on output, not judgment. It shows efficiency, not leadership. Amazon doesn’t need process tweakers — it needs principle enforcers.

GOOD: "Paused sprint changes after discovering team was already overloaded. Instead, ran a psychological safety pulse check and used results to renegotiate roadmap with director."

This shows diagnosis, constraint identification, and escalation — all Leadership Principles in action.

BAD: "Presented 90-day plan with 15 initiatives to my boss on day 5."

This signals arrogance and lack of learning humility. You can’t diagnose in five days.

GOOD: "Submitted a 2-page diagnostic memo at day 25 highlighting one systemic bottleneck and proposed experiment."

This shows patience, depth, and a preference for quality over quantity.

BAD: "Got positive feedback from my team in 1:1s."

This is vanity data. At Amazon, positive feedback without behavioral change is noise.

GOOD: "Three ICs independently adopted my decision log template within six weeks."

This shows influence, replication, and systems thinking.

More PM Career Resources

Explore frameworks, salary data, and interview guides from a Silicon Valley Product Leader.

Visit sirjohnnymai.com →

FAQ

What if my boss expects me to deliver fast results in the first 30 days?

Then your boss may not be operating at the Amazon bar. The Leadership Principle “Think Big” includes pacing — sustainable velocity over heroics. Push back with data: “Here’s what I can learn in 30 days, and here’s the risk of premature execution.” One L5 did this and was backed for a higher scope the next cycle. Speed without alignment is misalignment.

Should I follow a standard 30-60-90 template from the internet?

Not even close. Most templates are generic and fail Amazon’s “Bar Raiser” screen. They emphasize activity, not judgment. The ones we see in HC reviews that succeed are minimal: one clear constraint, one experiment, one escalation. If your plan looks like everyone else’s, it won’t move the needle.

How detailed should my 90-day plan be?

One page. Two pages absolute max. Amazon values conciseness as a proxy for clarity. If you need more than two pages, you haven’t made your trade-offs. The best 90-day plans I’ve seen were structured as: Problem → Hypothesis → Experiment → Success Signal → Risk. No timelines, no Gantt charts. Just judgment.