The first PIP is not a rescue plan. It is a proof test for whether you can manage standards without hiding behind empathy.
How to Handle an Underperformer at Amazon: First PIP Process for New Managers
TL;DR
The first PIP is not a rescue plan. It is a proof test for whether you can manage standards without hiding behind empathy.
In Amazon-style orgs, the problem is usually not that the employee missed once. The problem is that the misses became legible, repeatable, and expensive.
Treat the first PIP as a 30-day to 45-day management decision with weekly evidence, a written standard, and a clear end state.
Running effective 1:1s is a system, not a talent. The Resume Starter Templates includes agenda templates and question banks for every scenario.
Who This Is For
This is for a first-time Amazon manager, usually L5 or L6, who inherited 4 to 8 direct reports and now has one person slipping on output, judgment, or ownership.
It is also for the manager who still confuses discomfort with unfairness. If you came from an IC role where your own work was visible in a small number of launches or documents, the shift is brutal: now you are being judged on whether you can separate a fixable gap from a terminal one.
What actually matters when you decide to PIP someone?
The decision is about pattern, not mood. In a Q3 calibration I watched, the room did not care that the manager liked the employee; it cared that three written commitments had missed the bar in the same way.
Not a personality verdict, but a performance verdict. Not a punishment, but a threshold decision. That distinction matters at Amazon because managers often delay action while telling themselves they are being humane.
The real question is whether the gap is visible in the work. If the issue is one missed deadline with a clean explanation, you coach. If the issue is repeated misses after explicit expectations, the org stops treating it as a development problem.
Amazon culture makes this sharper because written evidence carries more weight than intent. A manager can say, “He is trying,” and still lose the room if the doc trail says the same defect returned three times.
The counter-intuitive part is that leniency often reads as weak judgment. In the debriefs I sat through, the manager who waited too long was rarely seen as kind. He was seen as someone who exported the cost of indecision to the team.
> 📖 Related: Google PM vs Amazon PM Total Comp: Level-by-Level Comparison (L3 to L7)
How do you tell coaching failure from real underperformance?
Real underperformance shows up after clarity. Coaching failure shows up when the manager never made the standard concrete in the first place.
In one skip-level conversation, a manager argued the employee “didn’t understand the bar.” HR asked for the first written expectation. There was none. That was not a PIP case; that was management debt.
Not vague feedback, but specific commitments. Not “communicate better,” but “send a same-day update when a dependency slips and escalate by noon.” If you cannot define the behavior, you cannot defend the PIP.
Look for repetition after correction. One miss is noise. Two misses after the same correction are signal. Three misses after the same correction are usually the org telling itself a story it does not want to admit.
This is where organizational psychology matters. People systematically blame character when structure is missing, and they blame structure when character is missing. Your job is to separate the two with dates, artifacts, and impact.
If the person improves when expectations are written down, the problem was often management ambiguity. If the person improves for a week and then returns to the same miss, the issue is usually not motivation. It is capability, judgment, or fit.
What do you say in the first conversation?
You say the gap plainly and stop talking. The worst first PIP conversations sound like apology speeches, and the employee leaves with sympathy but no understanding.
In a debrief after a failed first PIP, the hiring manager said the manager “kept softening every sentence.” That is a common mistake. Soft language does not protect the employee. It only protects the manager from hearing his own judgment.
Not a therapy session, but a business conversation. Not a threat, but a formal statement of current performance and expected change. The employee should leave knowing three things: what is wrong, what must change, and how long they have.
The best first conversation has four pieces. State the gap. State the business impact. State the support. State the timeline.
At Amazon, the conversation needs to sound calm and final, not dramatic. Dramatic language usually means the manager waited too long and is now trying to compensate with intensity.
A first PIP should not introduce surprise. If the employee is shocked, the manager failed earlier. The PIP is then just the first time honesty entered the room, which is already a managerial failure.
> 📖 Related: Google PM vs Amazon PM Interview: 5 Key Differences in 2026
How should you run the 30-day plan?
You run it weekly, in writing, with artifacts. A first PIP without written check-ins is theater.
In practice, a 30-day plan is usually enough to prove one of two things: the employee can close the gap, or the gap is deeper than hoped. Stretching that to 90 days often just drags the team through avoidable ambiguity.
Not more meetings, but more evidence. Not “how do you feel,” but “what shipped, what did not, and what was the outcome.” Amazon rewards the manager who can translate vague concern into dated proof.
Use a small number of measurable outcomes. Three is usually enough. More than that and the plan becomes noise. Fewer than that and the plan becomes a slogan.
In one manager discussion, the director asked for the last two work samples, the weekly recap notes, and the explicit success criteria. The manager had only memory. Memory loses to documentation every time.
The insight layer is simple: organizations do not remember performance the way people do. They remember the paper trail, the manager’s consistency, and whether the story can survive a skeptical reader in five minutes.
What does a defensible first PIP need to show HR and your director?
It needs to show a pattern, a standard, a response window, and a decision point. If it cannot survive a five-minute read from a director, it is too weak.
In a Q4 HC-style discussion, the question was not whether the manager was frustrated. The question was whether a peer manager, seeing the same record, would reach the same conclusion. That is the real bar.
Not perfect prose, but repeatable facts. Not emotion, but chronology. Not a vague summary, but the exact missed commitment, the date it was expected, and the business effect.
A defensible document usually includes three things: the observed gap, the explicit standard, and the support already given. If any one of those is missing, you have created a story, not a case.
This matters because first-time managers often over-index on fairness language. Fairness is not the issue. Legibility is the issue. If the case is legible, HR can work with it. If it is fuzzy, everyone gets cautious.
The practical judgment is blunt: if you are still arguing the case after the employee has already seen two clear warnings, you are late.
Preparation Checklist
Use this list before you open the PIP conversation. The process is simple only after the work is already done.
- Gather the last 3 examples of the performance gap with dates, deliverables, and impact.
- Write the expected standard in one sentence, not a paragraph.
- Align with your manager and HR before the employee hears anything.
- Set a 30-day or 45-day plan with weekly check-ins and one final decision date.
- Work through a structured preparation system (the PM Interview Playbook covers Amazon-style performance narratives, manager readouts, and calibration logic with real debrief examples).
- Decide what success looks like by day 10, day 20, and day 30.
- Decide now what you will do if the employee improves halfway and then regresses.
Mistakes to Avoid
The common failures are not subtle. They are predictable, and they are usually self-protection disguised as management.
- Turning the PIP into a character critique
BAD: “You need to be more positive and show better ownership.”
GOOD: “Your last three deliverables missed the agreed bar because the review inputs were incomplete at the checkpoint.”
- Using the PIP as a threat instead of a decision
BAD: “If this does not improve, HR will step in.”
GOOD: “Here is the gap, here is the support, here is the 30-day timeline, and here is the decision at the end.”
- Writing a vague improvement plan
BAD: “Improve communication and execution.”
GOOD: “Send same-day escalation notes, close weekly deliverables by the agreed date, and submit a written status update every Friday.”
FAQ
- Is a first PIP always a sign the person should leave?
No. A first PIP is a sign the current system has stopped working. Sometimes the employee recovers. Sometimes the manager discovers the gap is structural, not personal. The judgment is not about hope. It is about whether the work changes inside the stated window.
- Should a new Amazon manager wait to see if the problem resolves itself?
No. Waiting is usually just fear wearing a polite face. If the same miss repeats after explicit feedback and a written expectation, you do not need more optimism. You need a documented decision.
- What if I realize I was too vague earlier?
Own it immediately. That does not erase the employee’s gap, but it changes the frame from “surprise punishment” to “late correction.” If your expectations were fuzzy, the first repair is managerial: make the standard explicit before you ask for accountability.
Ready to build a real interview prep system?
Get the full PM Interview Prep System →
The book is also available on Amazon Kindle.