Performance Management

Performance Management: Meaning, Cycle, Models and Examples

Performance management is the continuous process of setting goals, giving feedback, assessing results and developing people so that individual work links to business outcomes. It runs all year through planning, monitoring, coaching, review and reward.  

A performance appraisal is only one event inside that cycle. Done well, it tells you who to promote, who to move internally and where you genuinely need to hire. 

What is a performance management system (PMS)? A performance management system, usually called PMS in Indian companies, is the set of processes, tools and rules an organisation uses to set goals (often as KRAs and KPIs), track progress, rate performance and link the rating to increments, variable pay, promotion and development. 

What is Performance Management? 

Performance management is the set of linked activities through which a company agrees what good looks like for each role, tracks progress against it, corrects course during the year, and uses the resulting evidence for pay, promotion, mobility and succession decisions. It is a management practice whose design HR owns, not an HR form that managers fill in. 

What it is not: 

• Not the annual appraisal. The appraisal is the review stage of a five-stage cycle. 

• Not a ranking exercise. Forced distribution is a calibration tool, not a purpose. 

• Not a performance improvement plan. A PIP is one response to a specific gap, used in a narrow set of cases. 

• Not an engagement survey. Engagement data explains why performance moves; it does not measure it. 

What a Performance Management System (PMS) Includes 

In Indian organizations, “PMS” usually refers to the whole system, not just the software. A working PMS has: 

• A goal framework (KRAs and KPIs, OKRs or a balanced scorecard) 

• A review cadence (check-ins through the year and a formal year-end review) 

• A rating scale with clear definitions 

• Calibration rules for consistency across managers and locations 

• A link from ratings to increments, variable pay and promotion (see pay for performance) 

• Development plans, an appeal process and a written policy 

• Usually, a software platform that holds all of the above 

Why Enterprise HR Teams Keep Rebuilding It 

Dissatisfaction is well documented. Gallup found that only 14% of employees strongly agree their performance reviews inspire them to improve, and only 29% strongly agree their reviews are fair. 

McKinsey frames performance management around four elements (goal setting, performance reviews, ongoing development and rewards) and argues that design choices across them should fit the organization rather than follow a template. 

For a CHRO running 8,000 people across tech delivery centres, bank branches and plants, the stakes are concrete. Poor ratings data means you can’t tell a real skills gap from a weak manager, so you hire externally for roles an internal candidate could have filled. 

Objectives of Performance Management 

State them explicitly, because systems drift towards whichever objective the loudest stakeholder cares about. 

• Align each role’s goals with the business plan, so effort lands where revenue does. 

• Give people frequent, specific feedback they can act on before the year ends. 

• Produce defensible evidence for pay, promotion and exit decisions. 

• Find successors and high-potential talent early enough to develop them. 

• Feed workforce planning: who can move, who must be hired, where capability is thin. 

Before you apply any rating scale, cadence or appeal rule from this page, check your own written performance policy. Summaries leave out the details that cause disputes, such as eligibility cut-offs, how probationers are rated, and the appeal window. 

KRAs and KPIs in Performance Management 

Most Indian goal sheets are built on KRAs (key result areas) and KPIs (key performance indicators). 

	KRA	KPI
What it is	A broad area of responsibility the role must deliver	A measurable indicator of performance within that area
Example (sales manager)	Revenue growth	Achieve ₹4 crore in new revenue this year
Example (HR business partner)	Talent retention	Keep regretted attrition below 10% in the business unit
Number per role	4–6	1–3 per KRA

A good goal sheet has a handful of weighted KRAs, each with one to three measurable KPIs, plus one or two behaviour measures. 

Performance Management vs Performance Appraisal 

Performance appraisal is a periodic evaluation that rates past work, usually once or twice a year, and feeds pay and promotion decisions.  

Performance management is the year-round system that contains it: goal setting, coaching, mid-cycle correction, development, and reward.  

Appraisal looks backward and judges; performance management looks forward and builds.  

Treating the two as synonyms is the most common design error in Indian enterprises. 

Dimension	Performance management	Performance appraisal
Time frame	Continuous, across the full year	A point-in-time event, usually annual or half-yearly
Primary purpose	Improve future performance and capability	Judge and record past performance
Owner	Line manager, with HR as system designer	Line manager, with HR as process administrator
Main output	Clearer goals, coaching, development plans, mobility decisions	A rating, a score, a documented record
Employee role	Active: sets goals, self-assesses, seeks feedback	Mostly passive: responds to the rating
Tools	KRAs/KPIs, OKRs, check-ins, 360-degree feedback, development plans, calibration	Rating scale, appraisal form, normalisation curve
Failure mode	Dies of neglect between cycles	Becomes a pay formula people game
Link to pay	Indirect and lagged	Direct and immediate

Read the table this way: if your only output at year-end is a rating, you have an appraisal, not performance management.  

The practical test is whether a mid-year course correction ever happens. If goals set in April are never touched until March, the system is a filing exercise. For appraisal techniques themselves, see our guide to performance appraisal methods. 

The Performance Management Cycle in Five Stages 

Five stages, with owners and timing that work for a large Indian organization on an April-to-March financial year. For a stage-by-stage deep dive, see the performance management cycle. 

1. Plan (April). Manager and employee agree goals cascaded from the business plan, plus behaviors or competencies. HR publishes the goal template and rating scale. Output: a signed, weighted goal sheet. Aim to have nearly all goal sheets locked within six weeks of the cycle starting. 

2. Monitor (May to March). Short monthly or fortnightly check-ins: what moved, what’s blocked, what changes. HR’s job is to make this cheap: a 20-minute conversation with three fields logged, not a form. This is the stage that collapses first when business gets busy. 

3. Develop (ongoing, with a mid-year review in September or October). Skills gaps found during monitoring become a performance development plan: training, a stretch assignment, a mentor or a lateral move. Mid-year is also when goals are rewritten if the business plan has changed; in a volatile year, a meaningful share of goals may need restating. 

4. Review (February to March). Self-assessment, then manager assessment, then calibration. Calibration sits between the manager’s rating and the final rating and is essential in a multi-location company. 

5. Reward and decide (March to April). Final ratings flow into increments, variable pay, promotion, succession slates and, where needed, a performance improvement plan. The same ratings should flow into workforce planning; most companies stop paying. 

The cycle only works if the monitoring stage is protected. A manager who skips check-ins from July to January arrives at review with no evidence and rates on recent memory, which is exactly how bias enters. 

The Five Core Components of Performance Management 

Goal Setting and Alignment 

Goals do two jobs: they direct effort, and they create a shared record of what was agreed. Cascade from the business plan, cap goals at five to seven per person, and weight them. A plant manager with eleven goals effectively has none. 

Practical rules we recommend: 

• Write at least one collaborative goal shared with another function, to stop silo optimization. 

• Weight the top goal at 25% or more, or the sheet becomes a list of chores. 

• Revisit weightings at mid-year rather than letting stale targets drag ratings down. 

• For frontline and shift-based roles, use three or four hard output measures plus one behavior measure. Nothing more. 

Continuous Feedback and Communication 

Feedback frequency is the lever with the best return for the effort. When reviews happen once a year, they rarely change behavior; Gallup’s finding that only 14% of employees strongly agree their reviews inspire them to improve is the clearest evidence. Regular conversations are what close that gap. 

Build it with a fixed fortnightly one-on-one, in-the-moment feedback on live projects, a structured 360-degree feedback round for people leaders once a year, and peer recognition that costs nothing to give. 

Performance Evaluation and Assessment 

Evaluation should draw on more than the manager’s memory: self-assessment, project outcomes, peer input for collaborative roles, and client feedback for client-facing roles. Keep the rating scale short. A five-point scale with clear definitions beats a seven-point scale nobody can explain. 

Before you change scales mid-cycle, check your policy document. Rating definitions, eligibility rules and appeal windows are usually written down and often ignored, which is where disputes start. 

Development Planning and Career Growth 

The development plan is where performance management earns its keep for retention. 

Each rating conversation should end in one of three outcomes: ready now for a bigger role; ready in 12 to 18 months with specified development; or needs to close a defined gap in the current role. “Keep up the good work” is a wasted cycle. 

Link the plan to real opportunities: a cross-functional project, a six-month rotation, a mentor two levels up, or a certification where the skill is technical and testable. 

Recognition and Rewards 

Rewards must be timely and explainable. Employees accept a modest increment far better than one they can’t trace to a decision rule. Pair money with non-monetary recognition and make some of it peer-nominated so it doesn’t all depend on one manager’s preferences. 

Choosing a Model: Continuous PM, OKRs, 9-Box Grid and Balanced Scorecard 

Pick the model that matches your operating rhythm and the strength of your managers, not the one with the best case studies. 

Model	Best fit	Review cadence	Effort to run	Main risk
Continuous performance management	Knowledge work, product and tech teams, fast-changing priorities	Fortnightly or monthly check-ins, light annual summary	High on managers, low on HR admin	Dies quietly when managers get busy; little audit trail
Objectives and Key Results (OKRs)	Companies needing visible alignment across functions; tech and digital units	Quarterly goals, weekly or fortnightly progress review	Medium, heavy in year one	Fused with pay, so people set soft objectives
9-box grid	Succession planning and high-potential identification	Annual, inside calibration	Medium; needs disciplined calibration	Potential judged subjectively; labels stick
Balanced scorecard	Manufacturing, BFSI and units with hard operational and customer metrics	Quarterly review, annual roll-up	High on data plumbing	Too many measures; becomes a report nobody reads

How to prioritize if you are choosing one: use a balanced scorecard at unit level if your business runs on operational metrics; add continuous check-ins at individual level, because that’s the engagement lever; use the 9-box only inside calibration for succession; and treat OKRs as optional unless cross-functional alignment is your named problem. Running all four at once is how you get process fatigue. 

Continuous Performance Management 

Replace the single annual conversation with monthly or quarterly check-ins and real-time goal adjustment. It suits teams whose priorities change quarterly.  

It needs capable managers, which is why it fails when rolled out before supervisors are trained. 

Objectives and Key Results (OKRs) 

Quarterly objectives, each with three to five measurable key results, visible across the company. The discipline is public progress scoring, not the goal sheet.  

One rule we’d enforce: keep OKR scores out of the increment formula. Once OKRs drive pay, ambition drops and everyone sandbags. 

The 9-Box Grid 

A three-by-three grid plotting current performance against assessed future potential. The cells run from low performance and low potential (bottom left) to high performance and high potential (top right), where the succession pool sits. 

Its value is forcing a conversation about potential that ratings alone never trigger. Its weakness is that potential is the softest judgement in HR.  

Define it with three observable criteria (learning agility, scope handled under pressure, and influence beyond the role) and require evidence for each, or the grid becomes a popularity map. 

The Balanced Scorecard 

Four perspectives: financial, customer, internal process, and learning and growth.  

For a manufacturing unit: cost per unit and margin; on-time-in-full delivery and complaint rate; cycle time and safety incidents; certification coverage and attrition in critical skills. Cap each perspective at three measures. 

Technology and AI in Performance Management 

Most enterprises now run performance management on a platform rather than spreadsheets, with goal tracking, continuous feedback, and analytics as core modules. 

What the platform genuinely fixes: 

• Goal visibility across levels, so cascading can be checked rather than claimed. 

• Feedback captured when it happens, the only reliable way to beat recency bias at review time. 

• Calibration support: rating distributions by manager, unit and location before ratings are locked. 

• Analytics linking ratings to attrition, promotion and internal movement. 

What it doesn’t fix: a manager who avoids hard conversations. Tooling turns a bad process into a faster bad process. 

Where AI Helps and Where It Introduces Bias 

AI earns its place in three narrow jobs: 

1. Summarizing a year of check-in notes into a draft review, which saves managers time and reduces recency bias. 

2. Flagging rating anomalies, such as a manager who has rated nobody below the midpoint for three cycles. 

3. Matching development needs to internal learning content and open internal roles. 

It is risky in three others: 

1. Scoring performance from activity data (emails sent, tickets closed, hours logged) measures visibility, not contribution, and penalises roles with long deliverables. 

2. Sentiment analysis of written feedback carries language bias. That matters in Indian workplaces, where feedback is often written in a second or third language and indirect phrasing is normal. 

3. Flight-risk prediction, if visible to managers, changes how they treat the flagged person, which can make the prediction self-fulfilling. 

Our rule: let AI draft and detect, never decide. Any model output that affects pay or exit needs a human reviewer who can explain the reasoning without referring to the model. 

Employee data: monitoring tools, activity data and AI models process employees’ personal data. Under India’s data protection framework, tell employees what is collected and why, collect only what the purpose needs, and keep it secure. 

Mobile access matters more in India than most vendor demos assume. Plant supervisors and branch staff won’t log into a desktop portal. If a check-in can’t be completed on a phone in a few minutes, adoption among frontline staff will stay low. 

How Performance Data Should Feed Workforce Planning and Hiring 

This is where most performance management systems stop short, and where the money is. Ratings get used for increments, then filed. They should also tell you how many people you need to hire, for which roles and in which locations. 

Calibrated ratings plus a potential assessment give you, for each business unit, a count of people ready now for the next level and a count ready in 12 to 18 months.  

Set that against your approved headcount plan and expected attrition. The gap is your external hiring requirement.  

Anything you can close internally is a role you don’t have to source, screen and onboard from scratch. 

Performance signal	What it tells workforce planning	Action
High performance, high potential, ready now	An internal successor exists for the next level	Slate internally; don’t open an external requisition
High performance, ready in 12 to 18 months	A successor exists but needs development time	Use a bridging hire or interim arrangement; start the development plan now
Strong performance, low interest in moving	Role stability, low backfill risk	Protect with retention action, not promotion
Low performance concentrated in one location or manager	Likely a management or capability issue, not a talent-supply issue	Fix the manager before the headcount
High attrition among top-rated people in one skill family	A pay, role-design or career-path problem	Expect sustained external hiring until the root cause is fixed

When this link is built properly, three things improve. Backfill forecasting stops being reactive, because you see the promotion chain before it moves. Senior roles fill faster, because internal slates are assessed against a known rating history instead of a cold search.  

And cost per hire falls on roles filled through internal mobility, since sourcing and agency spending disappear and onboarding is shorter. 

Two conditions make or break it. Ratings must be calibrated across units, or you are planning on noise. And the potential assessment must capture willingness to move, including location: a high-potential plant engineer who won’t relocate isn’t a successor for a role in another state. 

This is the lens we bring at Taggd: hiring demand is downstream of performance and mobility decisions, so the recruitment plan should be built from the same data that drives the rating cycle, not from a separate requisition queue. When talent intelligence, internal capability and external market supply sit in one view, a CHRO can decide role by role whether to build, borrow or buy. 

If your calibration is weak, don’t use ratings for planning yet. Fix calibration first. Planning on unreliable ratings is worse than planning on headcount alone. 

Calibration and Rating Distribution in Large Organizations 

Calibration is the session where managers defend their proposed ratings in front of peers and a business leader, and ratings are adjusted for consistency. In a company with 40 locations and 300 managers, it is the main defence against one manager rating generously and another harshly for identical work. 

How to run it: 

• Group by comparable role and level, not by reporting line. Comparing two branch managers is useful; comparing a branch manager with a data engineer is not. 

• Require three pieces of evidence for any top or bottom rating, cited aloud by the manager. 

• Show the distribution before and after the discussion, so the panel sees what moved. 

• Use guidance ranges, not hard quotas. A forced curve applied to a genuinely high-performing team destroys trust faster than any other single policy. 

• Document every change with the reason. This record is what defends the decision later. 

• For shift-based and frontline workforces, calibrate at supervisor-group level and lean more on output data, because managers observe each person less. 

Record the final distribution by unit, location, gender and tenure, and review it. If one plant rates 40% of people in the top band and a comparable plant rates 8%, the difference is management behaviour, not talent. 

Handling Underperformance and Performance Improvement Plans 

A performance improvement plan (PIP) is a time-bound, written plan used when performance falls short of a clearly stated standard and informal coaching hasn’t closed the gap. It states the gap, the required standard, the support provided, the review points and the consequence of no improvement. It typically runs 30 to 90 days, with fortnightly reviews. 

Use it when the gap is capability or consistency and the standard was communicated earlier. Don’t use it as a pre-exit formality, as a response to a single missed target, or where the real issue is role fit, team conflict or an unresolved health matter. Employees recognise a PIP issued in bad faith, and so do labour authorities and courts. 

In India, documentation carries the weight. Keep the goal sheet, check-in records, any written warnings, the PIP document with the employee’s acknowledgement, and the review notes.  

Confirmation and termination terms sit in the appointment letter, your HR policy and, for establishments covered by them, the standing orders under the Industrial Relations Code, 2020; the Model Standing Orders, 2026 apply to many establishments with 300 or more workers.  

Read those before acting and take legal advice on exits rather than relying on a general guide. 

Also check the appeal route. Many Indian enterprise policies give employees a defined window, often 7 to 15 days, to contest a rating or a PIP outcome. Honour it visibly. Systems survive on perceived fairness far more than on design elegance. 

Metrics That Show Performance Management Is Working 

Pick six to eight measures and track them by unit and location. The targets below are rules of thumb from enterprise practice, not industry benchmarks; set your own based on your baseline. 

Metric	What it tells you	Suggested starting target
Goal sheets locked on time	Whether the planning stage runs at all	95% within six weeks of cycle start
Check-in completion rate	Whether monitoring survives the busy season	80% of scheduled check-ins held
Rating distribution spread by unit	Calibration quality and manager bias	No unit far out of line with the company average in the top band
Regretted attrition among top-rated staff	Whether reward and development work	Below company-wide regretted attrition
Internal fill rate for next-level roles	Whether the system produces successors	Varies by function; track the trend
Promotion-to-performance correlation	Whether ratings drive real decisions	Strong and stable year on year
Employee perceived fairness score	The trust signal that predicts participation	Above 70% agreement
Manager effectiveness rating from direct reports	The capability of the people running the process	Tracked, with the bottom decile in training

Read the set together. High check-in completion with flat ratings and no internal movement means the conversations are happening and the decisions are not. For the financial view, pair these with revenue per employee trends, which show whether capability gains reach output. 

Common Failure Points in Large, Multi-Location Organisations 

The same breakdowns come up again and again in enterprise performance management. 

• Manager capability, not system design. The system is usually fine; the people running it have never been taught to give hard feedback. Train supervisors on three skills: setting a measurable goal, giving corrective feedback in under ten minutes, and running a development conversation. 

• Location drift. Head office runs the process properly; a plant 900 kilometres away treats it as a form. Fix it with calibration panels that cut across locations and distribution reports that name units. 

• Process weight. When the appraisal form runs to four pages and six approval steps, managers optimise for finishing, not accuracy. Cut fields until the form fits on one screen. 

• Pay taking over. If every conversation is about the increment, development stops being discussed. Separating the development review from the pay review by six to eight weeks helps more than any wording change. 

• Frontline exclusion. Shift workers, field staff and contract-adjacent roles often sit outside the system, which is where attrition is highest. Give them a simpler cycle: four output measures, a quarterly supervisor review and mobile capture. 

• Employee scepticism. Built up over cycles where ratings looked arbitrary. Rebuild trust by publishing calibration principles, honouring the appeal window and showing at least one visible case where a rating changed on evidence. 

• Policy drift. Rating scales, eligibility and appeal rules get amended and never communicated. Reissue the written performance policy at the start of each cycle and have managers confirm they’ve read it. 

FAQs 

Performance management stops being an HR ritual the moment its output drives hiring decisions. 

If you want to connect calibrated ratings and internal readiness to a sharper external hiring plan across your locations, talk to the talent intelligence and RPO team at Taggd about building that view.