Feedback Template for First-Time Managers Using Amazon Leadership Principles

Feedback Template for First-Time Managers Using Amazon Leadership Principles

TL;DR

First-time managers fail when they treat feedback as a soft conversation rather than a data-driven audit. High-performance feedback at Amazon is not about emotional alignment, but about identifying the specific Leadership Principle (LP) gap and providing the evidence to prove it. Effective managers who cannot tie a performance deficiency to a documented data are ignored during calibration.

Who This Is For

This is for the newly minted L6 or L7 manager at Amazon or a FAANG-level company who has transitioned from an individual contributor role and is now struggling to deliver critical feedback without damaging rapport. It is specifically for those who find themselves in a performance review cycle where their direct reports are surprised by their ratings because the manager lacked the courage or the framework to be precise in real-time.

How do I structure a feedback template using Amazon Leadership Principles?

The structure must lead with the specific Leadership Principle violated, followed by the data-backed observation, and ending with the expected behavioral change. In a calibration meeting I ran last year, a manager tried to tell the committee that a developer was not a team player; the committee rejected the feedback because it was a personality trait, not a performance gap. The problem is not the lack of kindness, but the lack of evidence.

A valid template follows a three-part logic: Principle, Evidence, and Delta. You start by naming the LP, such as Ownership or Insist on the Highest Standards. You then provide the evidence—not a feeling, but a specific document, a missed deadline, or a flawed architectural decision. Finally, you define the Delta: the exact difference between the current behavior and the expected L6/L7 bar.

The goal is to remove the manager's ego from the equation. When you use the LPs as the yardstick, you are not attacking the person; you are measuring the person against the company's stated values. This shifts the conversation from a subjective conflict to an objective gap analysis. It is not a critique of the individual, but a misalignment with the organizational standard.

Why is data more important than sentiment in manager feedback?

Data is the only currency that holds value during a talent review or a Performance Improvement Plan (PIP) discussion. I remember a debrief where a manager argued that an employee was lacking Dive Deep, but they couldn't produce a single instance where the employee missed a critical detail in a PRD. The result was a stalemate that delayed a necessary termination by three months.

Sentiment is an opinion; data is a signal. When a manager says a report is not taking enough ownership, that is sentiment. When a manager says the report missed three critical edge cases in the Q3 roadmap that resulted in a 4% drop in conversion, that is data. The difference is the difference between a productive coaching session and a HR grievance.

The organizational psychology here is simple: employees can argue with your perception, but they cannot argue with a log. By grounding feedback in the LPs, you are utilizing a shared language that the company has already validated. This prevents the employee from feeling singled out and forces them to engage with the business reality of their performance.

Which Amazon Leadership Principles are hardest for first-time managers to coach?

Ownership and Are Right, A Lot are the most difficult because they require judging a report's judgment, not just their output. In one specific case, a first-time manager struggled to tell a high-output engineer that they were failing Are Right, A Lot because the engineer was hitting all their deadlines but creating massive technical debt. The manager feared the engineer's productivity, ignoring the long-term cost.

Ownership is often confused with working long hours. A manager who rewards burnout instead of accountability is failing the LP. True ownership is not about the number of tickets closed, but about the willingness to dive into the root cause of a failure without being asked. Coaching this requires the manager to highlight the gap between doing the task and owning the outcome.

Are Right, A Lot is even more nuanced because it is about the quality of decision-making. First-time managers often fail here because they avoid challenging a report's logic. They mistake silence for agreement. To coach this, you must point to the specific decision point where the report failed to consider a critical variable, proving that their intuition was flawed in that instance.

How do I deliver critical feedback without losing the trust of my team?

You build trust through predictability and transparency, not through artificial positivity. I once saw a manager try to use the sandwich method—compliment, critique, compliment—only for the employee to completely ignore the critique and believe they were doing great. This creates a trust deficit when the employee eventually receives a low rating they didn't see coming.

The most trusted managers are those who provide feedback in real-time, often within 48 hours of the event. If you wait for the quarterly review to mention a failure in Bias for Action, you have failed as a manager. The report will feel blindsided, and the trust will evaporate. The signal must be immediate and tied directly to the LP.

The contrast is clear: trust is not built by avoiding conflict, but by removing the anxiety of the unknown. When a report knows exactly where they stand relative to the LPs every single week, they feel secure. They may not like the feedback, but they trust the process because it is consistent and evidence-based.

Preparation Checklist

• Audit the last 30 days of the report's output for specific evidence of LP alignment or misalignment.
• Map every piece of critical feedback to one of the 16 Leadership Principles (do not mix multiple LPs in one feedback point).
• Draft the Delta statement: clearly define the gap between current behavior and the expected level for their role (e.g., L5 vs L6).
• Review the feedback for subjective adjectives (e.g., "unprofessional," "lazy") and replace them with observed behaviors.
• Work through a structured preparation system (the PM Interview Playbook covers the Amazon-specific LP frameworks and how to map behaviors to levels with real debrief examples).
• Schedule the feedback session as a dedicated 1:1, not as a tack-on to a status update.
• Prepare the specific "What success looks like" criteria for the next 30 days to avoid ambiguity.

Mistakes to Avoid

Mistake 1: Using LPs as a weapon rather than a tool.

BAD: You are failing at Ownership because you didn't finish the report on time.

GOOD: On the Q3 Roadmap, the delivery was delayed by four days. This is a gap in Ownership because the risk was identified on day two but not escalated, resulting in a missed milestone.

Mistake 2: Generalizing behavior across the entire role.

BAD: You generally don't Dive Deep enough into the data.

GOOD: In the last three weekly business reviews, you were unable to explain the variance in the churn metric when asked. This indicates a need to improve your Dive Deep process for this specific KPI.

Mistake 3: Prioritizing the relationship over the bar.

BAD: I know you've been stressed, so let's just try to do better on the documentation next time.

GOOD: I recognize the workload is high, but the quality of the documentation does not meet the Insist on the Highest Standards bar. We need to find a way to maintain the bar while managing the volume.

FAQ

How often should I use LP-based feedback?

Weekly. Feedback should be a continuous stream of micro-adjustments, not a quarterly event. Waiting for a formal review to apply LPs is a failure of management that leads to attrition and poor calibration results.

What do I do if the employee disagrees with my data?

Move to a higher level of data. If they disagree with your interpretation of a document, bring in a third-party peer review or a technical lead to validate the standard. The debate should be about the data, not the opinion.

Can I use LPs for positive feedback?

Yes, and you must. Positive feedback tied to an LP reinforces the exact behavior you want to see scaled. Telling someone they did a great job is useless; telling them they demonstrated exceptional Bias for Action by resolving the outage in two hours provides a repeatable blueprint for success.amazon.com/dp/B0GWWJQ2S3).