---
name: analytics
description: Build HR analytics dashboards, workforce analytics models, predictive turnover models, headcount planning, people cost analysis, engagement dashboards, and org health metrics. Use when creating HR reports for leadership, forecasting staffing needs, analyzing compensation equity, measuring org health, or building people data infrastructure. Triggers on phrases like "HR analytics", "people analytics", "workforce dashboard", "turnover prediction", "headcount forecast", "org health", "compensation analytics", "people metrics", "HR reporting", "workforce planning data".
---

# HR Analytics

Transform people data into actionable insights for organizational decision-making.

## Workflow

1. Define the business question: what decision will this analysis inform?
2. Identify data sources: HRIS, ATS, payroll, performance system, survey tools, time tracking.
3. Clean and normalize data: handle missing values, standardize categories, deduplicate records.
4. Build analytical models or dashboards using appropriate methodologies.
5. Validate findings: check for statistical significance, bias, data quality issues.
6. Visualize results: create clear, actionable dashboards or reports.
7. Present to stakeholders with clear recommendations and next steps.
8. Establish ongoing monitoring and refresh cadence.

## Core HR Dashboards

### Executive People Dashboard

```
EXECUTIVE PEOPLE DASHBOARD — [Month/Quarter]
=============================================

HEADCOUNT OVERVIEW:
  Total employees: 1,245 (+3.2% QoQ)
  By department: Eng 412 | Sales 198 | Marketing 87 | Ops 245 | HR 32 | Finance 68 | Other 203
  By location: HQ 489 | Remote 512 | Satellite offices 244
  New hires (this quarter): 42    Separations: 28    Net: +14

TURNOVER:
  Overall voluntary turnover: 8.2% (annualized) — target: < 10%
  Involuntary turnover: 2.1%
  Critical role turnover: 5.4% — target: < 5%
  Top- performer turnover: 3.8% — target: < 3%
  First-year attrition: 12.3% — target: < 10%
  Top exit reason: Career opportunity (42%), Compensation (28%), Manager (15%), Other (15%)

ENGAGEMENT:
  Current eNPS: 38 (+4 from last quarter)
  Engagement survey score: 3.9/5.0 (+0.2 from last quarter)
  Active listeners (low engagement + still employed): 18% of workforce

DIVERSITY:
  Women in leadership: 34% (+2% YoY) — target: 40%
  Underrepresented minorities: 28% (+1% YoY)
  Pay equity ratio: 0.98 (women to men, adjusted for role/level)
  Disability representation: 4.2%

PEOPLE COST:
  Total comp cost: $72M (58% of operating expenses)
  Avg comp per employee: $57,800 (+5.1% YoY, inflation-adjusted: +2.1%)
  Turnover cost estimate: $1.8M (based on avg replacement cost of $64K)
  Training investment: $1.2M ($963/employee, 1.7% of payroll)

RED FLAGS:
  ⚠ Engineering retention dropped to 88% (from 93% last quarter)
  ⚠ Manager satisfaction in Ops department: 2.9/5.0 (lowest across org)
  ⚠ Time-to-fill for data roles: 68 days (exceeds 45-day target)
```

### Department-Level Dashboard

```
DEPARTMENT DASHBOARD — Engineering — Q1 2025
============================================

Headcount: 412 (target: 425, gap: 13 open requisitions)
Open reqs by level: Senior 5 | Mid 4 | Junior 3 | Principal 1

Turnover: 9.1% (company avg: 8.2%)
  By team: Platform 4.2% | Product 11.3% ⚠ | Infra 7.8% | QA 10.2% ⚠
  By tenure: <1yr 18.5% | 1-3yr 8.2% | 3-5yr 5.1% | 5+yr 3.4%

Engagement: 3.8/5.0 (company avg: 3.9)
  Low engagement drivers: "Insufficient growth opportunities" (32%),
  "Heavy workload" (28%), "Limited manager support" (18%)

Performance distribution:
  Top performer: 18% | Solid contributor: 65% | Developing: 14% | Underperforming: 3%

Internal mobility:
  Promotions: 8 (1.9%) | Lateral moves: 3 (0.7%)
  Promotion eligibility not promoted: 12 individuals (flag for review)

Recommendations:
  1. Investigate Product and QA team engagement drop
  2. Accelerate hiring for 13 open engineering positions
  3. Review promotion pipeline — 12 eligible candidates pending
  4. Launch engineering-specific development program
```

## Predictive Turnover Modeling

### Model Framework

```
PREDICTIVE TURNOVER MODEL
=========================

Target variable: Employee separation within next 6 months (Yes/No)
Training data: 3 years of historical employee records + outcomes
Model type: Logistic regression or random forest (explainable ML)

Features (predictors):
  Demographic: age, tenure, location, employment type
  Job-related: role, department, manager, reports count
  Compensation: salary vs. market ratio, time since last raise, bonus history
  Performance: last rating, rating trend, PIP history
  Engagement: latest survey score, eNPS response, participation rate
  Behavioral: PTO usage trend, overtime hours, meeting participation
  Career: time in role, promotion history, internal mobility activity
  Market: LinkedIn activity, job posting views (if available and compliant)

Output: Turnover probability score (0–100%) per employee
  > 70% = High risk (immediate intervention)
  40–70% = Moderate risk (monitor and engage)
  < 40% = Low risk (standard engagement)

Validation:
  - Train/test split: 80/20
  - AUC-ROC target: > 0.75
  - Precision-recall: optimize for precision (minimize false positives)
  - Fairness audit: ensure no adverse impact across demographic groups
```

### Intervention by Risk Level

```
HIGH RISK (> 70%):
  → HR business partner conducts stay interview within 1 week
  → Manager briefed (without disclosing model score)
  → Create personalized retention plan:
     - Career development conversation
     - Compensation review if below market
     - Workload assessment and adjustment
     - Manager coaching if relationship issue identified
  → 30-day follow-up check-in

MODERATE RISK (40–70%):
  → Manager conducts proactive 1-on-1 focused on development
  → Enroll in relevant L&D programs
  → Ensure inclusion in stretch assignments
  → Monitor engagement survey responses
  → Reassess risk score in 30 days

LOW RISK (< 40%):
  → Maintain standard engagement practices
  → Include in regular development planning
  → No specific action required
```

## Headcount Planning & Forecasting

### Forecasting Model

```
HEADCOUNT FORECAST — FY2026
============================

Starting headcount: 1,245

Planned additions:
  Organic growth (requisition pipeline): +85
  New initiative/hire freeze buffer: +15
  Total planned additions: +100

Projected separations:
  Voluntary turnover (8.2% annualized): -102
  Involuntary turnover (2.1%): -26
  Retirements (estimated): -8
  Total projected separations: -136

Projected ending headcount: 1,245 + 100 - 136 = 1,209

By department:
  Engineering: 412 → 425 (+13) — AI team expansion
  Sales: 198 → 210 (+12) — new territory coverage
  Marketing: 87 → 85 (-2) — efficiency optimization
  Operations: 245 → 260 (+15) — new product line
  HR: 32 → 36 (+4) — scale with org growth
  Finance: 68 → 70 (+2) — compliance expansion
  Other: 203 → 201 (-2) — shared services automation

Budget implications:
  Net headcount change: -36
  Estimated cost impact: -$2.1M (savings from natural attrition > planned hires)
  Average cost per new hire: $72K (including recruiting, onboarding, 1st year comp)
```

### Scenario Planning

```
SCENARIO: Revenue exceeds target by 20%
  → Accelerate engineering hiring: +25 additional roles
  → Expand sales team: +15 additional roles
  → Headcount budget adjustment: +$2.5M
  → Timeline: 6-month ramp, 9-month full capacity

SCENARIO: Economic downturn, 15% revenue reduction
  → Hiring freeze on non-critical roles
  → Voluntary separation program offered
  → Target headcount reduction: -8% (~100 FTE)
  → Retain top talent: identify and protect critical roles
  → Timeline: 3-month implementation, 6-month stabilization

SCENARIO: Major product pivot requiring new capabilities
  → Reskill existing workforce: 60% of gap through training
  → External hires: 40% of gap for unavailable internal skills
  → Headcount impact: neutral (reskill in place, targeted hiring)
  → Training budget increase: +$500K
```

## Compensation Analytics

```
COMPENSATION ANALYTICS DASHBOARD
================================

Overall compensation ratio (comp-to-market): 1.02
  Below market (< 0.90): 18% of workforce ⚠
  At market (0.90–1.10): 68% of workforce
  Above market (> 1.10): 14% of workforce

Pay equity analysis (controlled for role, level, experience, location):
  Gender pay ratio: 0.992 (women to men) — within acceptable range
  Ethnicity pay ratio: 0.987 — within acceptable range
  Unadjusted gap exists but narrows significantly with controls

Comp compression analysis:
  Cases where new hire earns more than incumbent: 12 cases (1.0% of workforce)
  Severity: 8 cases < 10% compression (acceptable), 4 cases > 15% (action needed)
  Recommended action: Accelerated review cycle for 4 compressed incumbents

Raise cycle effectiveness:
  Avg merit increase: 3.8% (budget: 4.0%, utilization: 95%)
  Top performer avg increase: 5.2%
  Mid performer avg increase: 3.5%
  Low performer avg increase: 1.0%
  Differentiation index: 1.7x (top vs. mid) — target: > 1.5x ✓
```

## Data Governance & Privacy

```
HR DATA GOVERNANCE PRINCIPLES
==============================

Data classification:
  PII (personally identifiable): Full name, SSN, address, ID numbers → Encrypted, role-based access
  Sensitive HR data: Salary, performance ratings, medical → Restricted access, audit logging
  Aggregated/anonymized: Dashboard data, trend analysis → Broad access

Access controls:
  HR analysts: Full access to analytical datasets (PII stripped where possible)
  Managers: Access to direct report data only (limited fields: name, role, performance status)
  Executives: Aggregated dashboards and reports only
  Employees: Self-service portal for own data

Retention policy:
  Active employee data: Retained indefinitely (updated regularly)
  Separated employee data: 7 years (legal requirement), then anonymized
  Survey data: 5 years, then aggregated and de-identified
  Performance data: 3 years, then archived

Privacy compliance:
  GDPR (EU employees): Right to access, right to erasure, data portability
  CCPA/CPRA (California): Consumer rights apply to employee data
  SOC 2 compliance: Annual audit of data security practices
```

## Edge Cases

- **Small organizations (< 50 employees)**: Focus on manual tracking and simple dashboards; predictive models need minimum 200+ data points
- **Rapid growth startups**: Prioritize headcount planning and turnover tracking; engagement data may be volatile
- **Multi-country operations**: Segment all analytics by country; account for local employment laws, cultural differences in survey responses
- **Post-merger integration**: Run parallel tracking for each entity; measure integration-specific metrics (retention of acquired talent, culture alignment)
- **Data quality issues**: Implement HRIS data validation rules; monthly data quality audits; assign data stewards per department