Performance improvement plan

A performance improvement plan (PIP) is a formal, time limited period in which one named employee is told exactly which standards they are missing, what support they will receive, and what will happen if the gap is still open at the end.

A PIP covers one person, not a service line. Wider delivery problems belong in service improvement work, while enterprise performance management handles the company level numbers. Mixing the three produces a plan nobody can act on.

It is also not a punishment, though it gets used as one. Written properly it is the record that the standard was stated, help was offered, and the employee had a fair run at closing the gap.

Key takeaways

• A PIP names one employee, the standard missed, the support offered and the date it is reviewed.
• The support half makes it defensible; a plan with no help attached rarely survives scrutiny.
• Standards have to be measurable before any plan can be written against them.
• The decision at the end goes in writing, whichever way it falls.

How it works

A PIP runs in five steps: state the gap against a written standard, agree the target, attach specific support, review on a fixed cadence, then decide. Every step produces a document, and the documents are the point.

Cadence is what separates a plan from a warning. Weekly contact keeps the evidence current and gives the employee somewhere to raise an obstacle before it becomes a reason.

Agree in advance what counts as sustained. One good week inside a bad month is not recovery — and saying so at the start avoids an argument at the end.

Public employers work to the same shape, and theirs is written down.

Under United States federal rules at 5 CFR Part 432, an agency facing unacceptable performance must give a reasonable opportunity to improve and must offer assistance to the employee in improving unacceptable performance.

That assistance wording was amended on 10 November 2022 — and it is the half most private employers leave out.

The standard being measured has to carry the weight. Title 5 of the US Code requires each agency appraisal system to set performance standards permitting accurate evaluation on objective criteria, as far as that is feasible.

Support usually means coaching rather than a training course, because the gap is nearly always situational rather than a missing qualification.

Name that support precisely. A promise of more coaching is not a commitment; two sessions a week with a named team leader, scheduled in advance, is one.

In contact centres that often takes the form of side by side coaching, where a team leader sits with the agent through live calls and corrects in the moment.

Write the target against a measure that already exists. A key performance indicators template stops managers inventing a new metric for the occasion, which is the fastest way to lose an appeal.

Examples

Three settings show the same process at very different levels of rigour: a United States federal agency working to codified rules, an outsourced contact centre in the Philippines, and a shared services finance team. Only the paperwork really changes.

United States federal agencies. The process is codified: tell the employee which critical element is unacceptable, state the standard that must be reached, offer assistance, and only then consider a reduction in grade or removal.

The rules also say the agency should tell the employee plainly that unless performance improves and stays improved, a reduction in grade or removal may follow. No surprises survive that wording.

Outsourced contact centres in the Philippines. An agent below a quality threshold moves onto a thirty day plan with weekly reviews and recorded call reviews attached.

Call centre performance evaluation supplies the score that plan is written against — which is why the scoring has to be stable before anyone is placed on one.

Shared services finance. An accounts payable clerk whose error rate sits above the team’s agreed ceiling gets a plan built around sampling, not opinion. Every weekly review cites a number, and the number comes from the same report each time.

Pick the sample size before the first review, so nobody can argue afterwards that one unlucky fortnight decided the outcome.

Related terms

• Service improvement: fixing a whole service line, where a PIP fixes one person’s output.
• Call quality monitoring: the routine scoring that usually produces the evidence a plan is built on.
• Coaching: the support activity attached to a plan, rather than the plan itself.
• Side-by-side coaching: live, in the moment correction, used heavily during contact centre plans.
• Call centre performance evaluation: the scoring method that defines what counts as missing the standard.

FAQ

How long should a performance improvement plan run?

Long enough for the employee to show sustained change on the measure in question, which in practice is usually 30 to 90 days. Anything shorter rarely produces evidence either side can rely on.

Does a PIP always end in dismissal?

No, and a process where it always does is not a plan. The point is a fair chance to recover, and plenty of employees finish one and stay. Managers who treat it as a formality tend to find that out at an appeal.

What must a plan contain?

The standard, the current result, the target, the named support, the review dates and the consequence. Leave out the support and you have a warning letter with a longer title.

Who should write it?

The line manager, with the human resources team checking the wording and the evidence. A plan the employee first reads on the day it begins has already lost half its usefulness — the measure should be familiar long before then.

What is the most common mistake?

Writing the plan against a standard nobody documented beforehand. Reviewers and tribunals both start there, and a vague standard collapses the rest.

Publish how your teams are coached and measured in the Outsource Accelerator provider hubs so buyers can see it.