At 80 employees, the founder knows who is carrying the company. At 800 employees, that memory breaks - and suddenly promotions feel political, managers avoid hard feedback, and high performers wonder why extra effort looks the same as average work.
Performance management is not an appraisal form; it is the operating system that aligns goals, feedback, rewards and development.
A growing company must move from founder judgement to transparent, manager-led, evidence-based performance cycles.
The core design is: strategy-linked goals, frequent check-ins, outcome-plus-behaviour evaluation, calibration, and development actions.
Do not copy a large-company bell curve blindly; choose a system that fits business stage, role type and culture.
Track success using check-in coverage, goal completion, rating skew, regretted attrition, internal mobility and improvement closure.
The best answer separates performance, potential and pay - they are related, but not the same decision.
Big Picture: From Annual Judgement to Continuous Performance
A scaling company outgrows informal performance calls because complexity rises faster than leadership visibility. The redesign is a shift from one annual judgement event to a repeatable system that makes expectations visible, feedback timely and decisions defensible.
The central shift is from judging people once a year to managing performance throughout the year.]
<h2>Core Explanation: How to Redesign the Performance System</h2>
<p>The big idea: a good performance management system answers four questions clearly for every employee - <strong>what matters, how am I doing, what will happen because of it, and how do I improve?</strong></p>
<p>In a growing company, the redesign must balance two forces. It needs enough structure to be fair across teams, but enough flexibility to handle different roles - sales, product, engineering, operations and support cannot be measured identically.</p>
<roadmap-steps
data-steps='[
{"title":"Start with business strategy","desc":"Translate the company's growth priorities into team and individual goals, such as revenue quality, retention, speed, reliability or customer experience."},
{"title":"Define role-level outcomes","desc":"Clarify what good performance means in each role using KPIs, OKRs or role scorecards, not vague traits like ownership."},
{"title":"Build a feedback cadence","desc":"Replace surprise annual reviews with monthly or quarterly check-ins, documented coaching and mid-cycle course correction."},
{"title":"Evaluate outcomes and behaviours","desc":"Measure both what was achieved and how it was achieved, especially collaboration, ethics, customer orientation and leadership behaviours."},
{"title":"Calibrate and decide consequences","desc":"Use cross-manager calibration to reduce rating inflation, then link outcomes to rewards, development, promotions or performance improvement plans."}
]'>
</roadmap-steps>
[[FIGURE: {"layout":"cycle","items":[{"label":"Set Goals","note":"Clear outcomes"},{"label":"Check In","note":"Coach early"},{"label":"Review Evidence","note":"Outcomes plus behaviour"},{"label":"Calibrate","note":"Reduce bias"},{"label":"Develop","note":"Grow capability"}]} | caption: Performance management is a cycle, not a once-a-year HR ritual.]
<h2>The Three Design Choices That Make or Break the System</h2>
<h3>1. Goal Architecture: What Should People Be Measured On?</h3>
<p>Use <strong>OKRs</strong> for ambitious, cross-functional goals and <strong>KPIs</strong> for recurring operational performance. For example, a product manager may have OKRs around adoption and customer outcomes, while a customer support lead may track resolution time, quality scores and escalation rate.</p>
<tip-box data-type="info" data-title="Example - Indian Growth Company" data-icon="📌">
<p>In Indian startups such as Razorpay, Swiggy or Urban Company, teams often work across product, operations, compliance and customer experience. A performance system must therefore convert company priorities into team-level measures rather than rely only on manager impressions. The strategic so what: scale requires translation of strategy into measurable local ownership.</p>
</tip-box>
<h3>2. Evaluation Logic: Outcomes Plus Behaviours</h3>
<p>A common trap is rewarding only the number. That can create short-term wins and long-term damage - sales that churn, code that breaks, or operations speed that hurts quality. Strong systems evaluate both <strong>results</strong> and <strong>values-aligned behaviours</strong>.</p>
[[FIGURE: {"layout":"matrix","xAxis":"Behaviour and Values","yAxis":"Business Outcomes","items":[{"label":"Role Model","note":"High results, right way"},{"label":"Risky Star","note":"High results, wrong way"},{"label":"Culture Carrier","note":"Low results, right way"},{"label":"Low Fit","note":"Low results, wrong way"}]} | caption: The best systems distinguish high performers from people who deliver numbers by damaging the culture.]
<h3>3. Decision Separation: Performance, Potential and Pay</h3>
<p><strong>Performance</strong> is about current contribution. <strong>Potential</strong> is about readiness for larger future roles. <strong>Pay</strong> is a reward decision influenced by performance, market compensation, affordability and retention risk. Mixing all three into one conversation makes managers defensive and employees confused.</p>
<data-table
data-headers='["Decision", "Question It Answers", "Typical Evidence"]'
data-rows='[
["Performance", "How well did the employee deliver in the current role?", "Goals achieved, quality, stakeholder feedback, behaviour examples"],
["Potential", "Can the person handle a bigger or more complex role?", "Learning agility, scope handled, leadership signals, problem-solving under ambiguity"],
["Pay", "What reward is fair and affordable?", "Performance, market benchmark, internal equity, budget, retention risk"]
]'>
</data-table>
<h2>Metrics to Track After the Redesign</h2>
<p>If HR cannot measure whether the new system works, managers will treat it as another process rollout. Use a small dashboard that checks adoption, fairness, development and business impact.</p>
<data-table
data-headers='["Metric", "Formula or Definition", "What Good Looks Like"]'
data-rows='[
["Check-in coverage", "Employees with documented check-ins ÷ total employees × 100", "Typical range is 0-100%; above 90% is strong if check-ins have usable notes."],
["Goal completion rate", "Completed goals or key results ÷ committed goals or key results × 100", "For ambitious OKRs, 60-80% is often healthy; around 70% with learning is strong."],
["Rating distribution health", "Percentage of employees in each rating band by team, manager and level", "Typical range is 0-100% per band; strong means no unexplained manager-level skew after calibration."],
["Regretted attrition", "High-performing voluntary exits ÷ total high performers × 100", "Typical range varies by industry; strong is below the company baseline and trending down."],
["Internal mobility rate", "Roles filled by internal moves ÷ total roles filled × 100", "Typical range depends on growth stage; a rising rate is strong when role quality is maintained."],
["Improvement closure rate", "Completed development or performance improvement actions ÷ actions opened × 100", "Typical range is 0-100%; above 80% is strong when outcomes are documented."]
]'>
</data-table>
<h3>Mini Worked Example: Is the New System Working?</h3>
<p>Assume a 600-employee company pilots the redesigned system for one quarter. These are hypothetical numbers, but the logic is exactly how you would reason in an HR case.</p>
<data-table
data-headers='["Measure", "Calculation", "Interpretation"]'
data-rows='[
["Check-in coverage", "540 employees with check-ins ÷ 600 employees × 100 = 90%", "Adoption is strong enough to judge the pilot seriously."],
["Goal completion", "420 completed key results ÷ 600 committed key results × 100 = 70%", "Healthy for ambitious OKRs if missed goals produced learning."],
["Regretted attrition", "6 high performers exited ÷ 120 high performers × 100 = 5%", "Needs comparison with past baseline; if earlier was higher, the redesign may be helping retention."],
["Improvement closure", "32 closed actions ÷ 40 opened actions × 100 = 80%", "Managers are not just rating people; they are following through on development."]
]'>
</data-table>
<h2>Definitions You Should Be Able to Say Cleanly</h2>
<tip-box data-type="info" data-title="Canonical Definition" data-icon="📘">
<p>Herman Aguinis defines performance management as “the continuous process of identifying, measuring, and developing the performance of individuals and teams.”</p>
</tip-box>
<check-list
data-items='[
"Performance appraisal: a periodic assessment of an employee's past performance, usually for documentation, reward or promotion decisions.",
"Performance management: the continuous system of goal-setting, feedback, review, development and consequences.",
"OKR: an objective states what matters; key results define measurable proof of progress.",
"Calibration: a cross-manager review that improves consistency and reduces rating inflation, severity bias and team-level unfairness.",
"Potential: a judgement of future role readiness, based on learning agility, leadership signals and ability to handle larger scope."
]'>
</check-list>
<h2>Case Study: Urban Company and Performance Management at Marketplace Scale</h2>
<tip-box data-type="info" data-title="Case Study - Urban Company" data-icon="🏆">
<p>Urban Company shows how a growing Indian platform uses training, standards, customer feedback and performance consequences to manage quality at scale.</p>
</tip-box>
[[GOLD-IMAGE: A smartphone on an apartment doorstep beside a toolkit and cleaning supplies, with a calm blue-green service checklist interface on the screen, no logo and no readable text | caption: Urban Company makes performance visible by turning service quality into repeatable signals across thousands of home visits.
Situation: Urban Company operates in home services, where the customer experience is created inside a customer's home by distributed service professionals. As the company scaled across Indian cities and categories such as beauty, repairs and cleaning, informal quality control would not be enough. One poor visit could damage trust in the entire platform.
The move: The company built performance management into the marketplace itself. The primary driver was standardisation of service quality through training, category-specific processes and customer feedback. Supporting drivers included app-based work allocation, visible ratings and reviews, service protocols, quality audits, partner support and consequences for repeated poor performance.
The lesson: Urban Company is not just “rating workers.” It is designing a performance system where expectations, measurement, coaching and consequences are embedded into daily work. The caution is equally important: platform performance systems must remain transparent and fair, because opaque algorithms or one-sided penalties can reduce trust among service partners.
So what for interviews: In a growing company, performance management must move closer to the work itself. The best systems do not only evaluate people after the fact; they create conditions for good performance before the work happens.
How AI Changes Performance Management
AI does not remove the manager from performance management. It changes what evidence managers can see, how quickly they can coach, and how carefully HR must govern bias.
Use NotebookLM: upload this lesson, the company's careers page and a recent annual report or investor update. Ask: “What business priorities should shape performance management here, and what interview questions could HR ask about redesigning it?”
Interview Relevance
“A 700-employee startup has grown fast, but employees complain that performance reviews are inconsistent and promotions feel unfair. How would you redesign the performance management system?”
Use the phrase: “I would redesign this as a performance operating system, not as a new appraisal form.” It signals maturity immediately.
Common Mistake
The single biggest mistake is redesigning only the rating form. That fails because the real problems are usually unclear goals, weak feedback, manager inconsistency and unfair consequences. One-line fix: start with business priorities and redesign the full cycle - goals, check-ins, evidence, calibration, rewards and development.
What to Revise Next
This is a natural capstone topic because it connects HR strategy, rewards, talent development, analytics, culture and manager capability. As your final review, pick one company you know well and practise a 6-minute answer that redesigns its performance system using the full cycle above.
Mark Lesson Complete (Redesigning Performance Management for a Growing Company: Interview Case Guide)
