Performance Improvement Plans Using Productivity Data: A Manager's Guide

Performance Improvement Plans are emotionally loaded for everyone involved. Productivity data — used correctly — turns the conversation from "I think you're not doing enough" to "here are the specific outputs, here is what good looks like, here is the support we'll give you." Used incorrectly, the same data turns the PIP into a labour-court matter.

This post is the manager's playbook for using productivity data in PIPs in a way that is fair, evidence-based, and defensible.

The 4-stage PIP structure

Stage 1: Signal (2-4 weeks)

Productivity dashboards surface a trend. Output is below team baseline; quality is dropping; response times are slipping. The data is the signal, not the verdict.

Before any conversation: cross-check the data. Is the metric measuring what you think it is? Was there a sick week? A project change? A tooling issue? Three rounds of "data + reality check" before going to stage 2.

Stage 2: Informal conversation (1 week)

One-on-one with the employee. Open with concern, not accusation:

"I've noticed your output on [specific deliverable type] has been about [number] over the last [time period], down from your usual [baseline]. I want to understand what's going on before assuming."

Listen for: workload pressure, personal circumstances, tooling problems, manager-side issues (unclear expectations, conflicting priorities). Most "performance" issues resolve at this conversation alone.

Stage 3: Formal PIP (4-12 weeks)

Only if Stage 2 doesn't resolve. The PIP document needs five things:

• Specific behaviours / outputs that need improvement — with measurable baseline and target
• Support the company will provide — training, coaching cadence, mentor, tooling
• Timeline — typically 30 / 60 / 90 days with checkpoint reviews
• Consequences of non-improvement — explicit, not euphemistic
• Employee's right to respond — in writing, before the plan is finalised

Stage 4: Review and outcome

At each checkpoint and at the end of the PIP, document objectively:

• Did the employee meet the targets? Specific numbers, not "improved a bit."
• If yes — exit the PIP cleanly with written confirmation. Do not roll into permanent suspicion.
• If no — proceed to separation or extension based on the original plan's stated consequences.

The three data mistakes that create legal exposure

Mistake 1: Using "activity time" as the metric

"You were active in apps for only 4 hours yesterday" is not a defensible PIP basis. An employee can be productive in 4 active-hours or unproductive in 9. The metric must be output (tickets closed, code merged, calls handled at quality), not activity.

Mistake 2: Selecting a comparison baseline that isn't fair

"You're below the team average" — but team average includes 3 people who joined last year doing different work. Compare like-for-like or against the employee's own historical baseline. Indian labour courts have repeatedly struck down PIPs based on inconsistent comparisons.

Mistake 3: Cherry-picking time windows

Pulling data from the worst 4 weeks of the year and presenting it as the trend. The defensible window is 8-12 weeks, looking at the moving average, with anomalies (sick leave, project change) explicitly noted and excluded.

Sample PIP framing

Good framing reads as evidence-based, not punitive:

"Over the past 8 weeks, your average tickets-resolved-at-quality has been 12 per week. The team average over the same period is 22 per week. Your own baseline from Q1 of this year was 24 per week. We expect tickets-resolved-at-quality of at least 20 per week by the end of this 60-day improvement plan, with a checkpoint review at 30 days. We will provide weekly coaching with [mentor name] for the duration. If you are not at 20 by day 60, the next conversation will be about role fit."

Notice what this framing includes: specific number, fair comparison (team and own history), concrete target, realistic timeline, support offered, explicit consequence.

What to never put in writing

• Subjective character judgements ("not committed," "lazy attitude") — these become evidence of bias if challenged
• References to monitoring data without explaining what was measured and how
• Comparisons that aren't quantifiable ("everyone else does more")
• Promises you cannot keep ("if you improve we'll definitely promote you")
• Anything related to protected characteristics (age, religion, caste, gender, family status, disability)

The conversation you don't have on the PIP

The single highest-leverage manager skill in PIPs is recognising when the productivity issue is actually a manager problem, a process problem, or a fit problem rather than an effort problem. Productivity data helps you distinguish:

The Indian-law context

Indian labour-law expects "natural justice" in performance management. The case law standard is roughly:

• Employee was given clear notice of expectations
• Employee was given opportunity to be heard
• Comparable employees were treated comparably
• The decision-maker was not biased
• The action was proportionate to the issue

A well-documented 4-stage PIP satisfies all five tests. Verbal informal warnings followed by sudden termination do not — and have repeatedly been overturned in Indian labour tribunals.

FAQ

Should productivity data be shown to the employee during the PIP conversation?

Yes. Full transparency is both fair and legally defensible. Walk through the dashboard with them; let them see exactly what you saw.

What if the employee disputes the data?

Treat the dispute as a separate process. Investigate (sometimes monitoring data has bugs, scope issues, or context the employee can explain). If the data is correct, document the explanation given.

How does this work for sales / commission-based roles?

Sales roles have clearer output metrics (revenue, pipeline) than activity metrics — making PIPs more straightforward. The activity data is diagnostic for sales teams, not the primary input.

Can monitoring data be used to defend the company in a wrongful-termination claim?

Yes, provided (a) the monitoring was consented to, (b) the data is preserved unmodified, and (c) the data was used consistently across employees. See our consent form template for the legal basis.

How does Headx help with PIP-relevant reporting?

Headx surfaces output-relevant metrics (productive-app time, task-completion via integrations, attendance) in employee-level reports designed for manager review. The reports exclude the metrics — keystroke counts, mouse movement — that should not be in PIPs.