Amazon PIP: Focus, Pivot, severance and the appeal

Every figure here is reported, not official

Amazon does not publish this process. What follows is assembled from journalism, plaintiff-side law firms, coaching firms with large Amazon caseloads, and employee accounts. Program names and severance numbers change, and yours may differ.

Treat the shape as reliable and the numbers as indicative. Nothing here is a promise about what you will be offered.

The reported pipeline

A negative Forte review or a "Needs Improvement" rating leads to Focus, then to Pivot, then to an outcome.

Focus is a coaching plan, reported at roughly one to three months, during which performance is closely monitored and heavily documented. Two edges are worth knowing:

• You may not be told you are in Focus unless you ask.
• Internal transfers stop being ordinary. Both the employee-side firm and the coaching firm this page rests on describe VP-level approval rather than a prohibition — Steller Law says Amazon "generally restricts employees in the Pivot Program from applying for internal transfers without VP level approval on both teams", and Rora says Focus "makes it much harder to switch teams (requires VP approval)". An earlier version of this page said transfers were blocked. That was wrong in the direction that matters, because it talks people out of the one move that sometimes works.

Leaving Amazon during Focus is reported to still count as unregretted attrition, which matters for the section below.

Reported durations, from the coaching firm's own account: Focus runs 60 to 90 days, and it was formerly called the Dev List, which is the name older threads use.

Pivot is the formal moment, and it presents three doors:

• Leave now with severance, in exchange for a release of claims
• Accept the improvement plan — specific deliverables over 30 or 60 days
• If the manager rules the plan failed, appeal to a panel of peers

Five days, reportedly, to choose between the three

The deadline is the part of this process people find out about last, and it is short. The employee-side firm that handles these cases reports that Amazon "typically provides a five-day window to decide" between taking the severance, entering the plan, and going to the panel.

Five days is not enough time to do the thing the decision actually requires, which is to get the release of claims in front of an employment lawyer and to work out what the alternative is worth. So the useful preparation happens before the meeting rather than after it, and this is the argument for reading a page like this one while nothing has happened yet.

Three things worth having ready in advance, none of which needs anybody's permission:

• A dated list of what you have delivered, running back over the last two review periods, written in the language your goals were written in.
• Your own copy of anything the company can edit. Review text, goal documents, written manager feedback. Saved to a personal device, with the date you saved it.
• The name of an employment lawyer you could call on a Tuesday. Finding one inside a five-day window, while deciding, is the version of this that goes badly.

If the window has already started, the first email to send is the one asking for the offer and the plan in writing, and asking what the deadline is. Both are reasonable requests, and both start a written record of what you were given and when.

The ladder declines at every stage

The reported severance figures cluster around two months' gross pay to leave at Pivot entry, roughly half that after failing the plan, and a small residual after losing an appeal. The exact numbers vary by case and by year.

The shape is the constant, and the shape is the point. The structure prices the decision: every door forward costs money, and the cost falls the longer you stay in the process. That is worth knowing on day one rather than at the end, because it means the decision has a clock on it independently of whether the plan is winnable.

What the appeal hearing is actually like

The page has described the appeal as a panel deciding a question. The firm's account is more specific, and the difference changes how you would prepare: both the employee and their manager present their cases, usually by videoconference, to a panel of peers, which then decides whether you continue in the Pivot program or are terminated.

So it is an adversarial hearing with two presentations, not a review of a file by people who have already read it. Three consequences follow:

• You will speak, and you should plan what you say. A presentation that walks the panel through the goals in order, with what was delivered against each, is a different artefact from a folder of evidence.
• Your manager presents too, which means the case you are answering is the one they make on the day rather than only the one already written down. Anticipating it is part of the preparation.
• The panel are peers at your level or above, so the register is a work conversation and not a legal one. What persuades is specificity about the work.

The question in front of them is a single one: did this person meet expectations.

Wins happen, and there are first-hand accounts of them. They are reported as uncommon, largely because by the time an appeal is heard the documentation file is deep. The cases that succeed lean on written peer feedback and on recorded manager statements that expectations were being met.

Which tells you what to assemble and when, if you intend to appeal: a dated list of deliverables mapped one to one against the Pivot goals, plus written statements from peers. Collected during the plan, not after it — by the hearing, a request for a supportive statement is a request made of somebody who has just watched what happens to people on plans.

Unregretted attrition is reported as a target, so a placement may say little about you

URA is a reported internal metric for departures the company does not mind, with team-level targets and a company-wide rate reported at around 6%. Plaintiff-side counsel argue the whole structure lets Amazon frame exits as resignations, through the severance door, or as independent panel decisions, through the appeal.

Read carefully, the quota is the most reassuring fact available to somebody in this process. If a target exists at all, then a placement may say very little about the person placed. That does not tell you which your case is. It does mean that reading a Pivot as a verdict on your ability is reading more into it than it can carry.

Two decisions people get wrong early

Do not sandbag or vent during Focus. Every one-to-one is being logged by the other side, and Focus is the stage where the record is thinnest and most changeable.

Do not choose a Pivot door on day-one emotion. The comparison is between the severance difference, a realistic view of the plan and the appeal, and — if you hold a work visa — the clock that starts on your last day of employment regardless of which door you take.

If you are on a work visa

Taking the voluntary severance door means leaving voluntarily, which can complicate US unemployment eligibility — the quit-versus-fired trap, in its most concrete form.

Separately, if you are on an H-1B or similar, none of this has any immigration effect until employment ends. Then a hard clock starts. The visa page covers it, and it is the one place where a later termination date can be worth more than a larger payment.

Before you sign a release of claims

Fairhanded is not a law firm, is not affiliated with Amazon, and has no access to its internal policies. Before signing a release of claims, take it to an employment lawyer — that document is where the money and the rights actually change hands.