| name | hr-analytics |
| description | Implement HR analytics and people analytics including predictive modeling, workforce metrics, turnover prediction, hiring analytics, performance forecasting, and data-driven HR decision-making. Use for: building HR dashboards, implementing predictive analytics, forecasting turnover, analyzing workforce trends, measuring HR effectiveness, creating people analytics strategies, and leveraging data for talent decisions. |
HR Analytics
Leverage data, statistical methods, and predictive modeling to transform HR from reactive administration to proactive, strategic workforce management.
Overview
This skill provides comprehensive frameworks for HR analytics, people analytics, predictive modeling, and data-driven decision-making. It covers descriptive, diagnostic, predictive, and prescriptive analytics, key workforce metrics, analytics tools, and implementation strategies.
Analytics Type Selection
| Scenario | Analytics Type | Purpose | Reference |
|---|
| Understanding what happened | Descriptive Analytics | Summarize historical HR data | /references/hr-metrics.md |
| Understanding why it happened | Diagnostic Analytics | Investigate root causes | /references/people-analytics-frameworks.md |
| Forecasting future outcomes | Predictive Analytics | Anticipate trends and risks | /references/predictive-modeling.md |
| Recommending actions | Prescriptive Analytics | Guide HR strategy and decisions | /references/people-analytics-frameworks.md |
| Building analytics capability | Implementation Strategy | Develop tools, skills, processes | /references/analytics-tools.md |
Four Types of HR Analytics
1. Descriptive Analytics
What: Summarizes historical HR data
Purpose: Understand what has already happened
Examples:
- Past turnover rates by department
- Historical performance review scores
- Headcount trends over time
- Diversity demographics
Tools: Dashboards, reports, data visualization
Value: Baseline understanding, trend identification
2. Diagnostic Analytics
What: Investigates reasons behind past events
Purpose: Understand why something happened
Examples:
- Why did turnover increase in Q3?
- Why is engagement lower in engineering?
- What factors correlate with high performance?
- Why are certain demographics underrepresented?
Tools: Root cause analysis, correlation studies, drill-down reports
Value: Actionable insights into problems
3. Predictive Analytics
What: Forecasts future trends and outcomes
Purpose: Anticipate what is likely to happen
Examples:
- Which employees are at risk of leaving?
- What are future hiring needs?
- Who are high-potential future leaders?
- What will engagement be next quarter?
Tools: Statistical modeling, machine learning, regression analysis
Value: Proactive interventions, strategic planning
4. Prescriptive Analytics
What: Recommends specific actions
Purpose: Guide HR strategy and decision-making
Examples:
- Optimal compensation adjustments to reduce turnover
- Targeted retention interventions for at-risk employees
- Personalized development plans based on career trajectory
- Ideal candidate profiles for hiring
Tools: Optimization algorithms, simulation models, AI recommendations
Value: Data-driven action plans
Key Predictive Analytics Use Cases
1. Turnover Prediction
Goal: Identify employees at risk of leaving
Data Sources:
- Tenure, performance ratings, compensation changes
- Manager relationship scores, engagement survey results
- Promotion history, training completion
- External factors (market conditions, commute)
Model: Logistic regression or decision tree
Output: Probability score (0-100%) that each employee will leave in next 6 months
Action: Targeted retention interventions (raises, promotions, role changes, manager coaching)
Example: HP's "Project Insight" reduced turnover from 20% to <10%, saving $300M
2. Hiring Success Prediction
Goal: Forecast candidate's future performance and fit
Data Sources:
- Resume patterns, assessment results, interview scores
- Education, skills, experience
- Hiring source, referral status
Model: Machine learning classification
Output: Predicted performance rating, time-to-productivity, retention likelihood
Action: Better hiring decisions, reduced mis-hire costs
Example: Google's "Prediction Engine" analyzes job history, education, skills, personality to predict success
3. Performance Forecasting
Goal: Predict future employee performance
Data Sources:
- Historical performance ratings, goal achievement
- Skill assessments, training completion
- Peer feedback, manager effectiveness
Model: Regression or neural network
Output: Predicted future performance rating
Action: Targeted development, succession planning, resource allocation
4. Workforce Planning
Goal: Anticipate future talent needs and skill gaps
Data Sources:
- Business growth projections, attrition trends
- Internal mobility patterns, market talent availability
- Skills inventory, training data
Model: Time series forecasting, scenario modeling
Output: Future headcount needs by role, skill gaps
Action: Proactive hiring, reskilling programs, strategic talent acquisition
Example: Cisco uses predictive analytics to anticipate skills gaps and address proactively
5. Engagement and Productivity
Goal: Forecast engagement levels and productivity trends
Data Sources:
- Engagement surveys, performance metrics
- Collaboration patterns, work-life balance indicators
- Manager effectiveness, team dynamics
Model: Regression, sentiment analysis
Output: Predicted engagement scores, productivity trends
Action: Early intervention to boost engagement, optimize team composition
Example: Best Buy linked 0.1% engagement increase to $100K revenue increase per store
Critical Workforce Metrics
Talent Acquisition
- Time-to-Hire: Days from posting to offer acceptance (Target: <30 days)
- Cost-per-Hire: Total recruiting costs / hires
- Quality of Hire: New hire performance after 6-12 months (Target: 4.0/5.0+)
- Offer Acceptance Rate: % of offers accepted (Target: 85%+)
- Source of Hire Effectiveness: Performance and retention by source
- Time-to-Productivity: Days until full productivity
Retention and Turnover
- Voluntary Turnover Rate: (Voluntary departures / avg headcount) x 100 (Target: <10%)
- Involuntary Turnover Rate: Performance management effectiveness indicator
- Turnover by Segment: Department, role, tenure, performance, manager
- Regrettable vs. Non-Regrettable: % high performers vs. low performers leaving
- Retention Rate: (Employees remaining / starting headcount) x 100 (Target: 90%+)
- Cost of Turnover: Recruiting + onboarding + lost productivity (50-200% of salary)
Performance
- Goal Achievement Rate: % of goals/OKRs met (Target: 70-80% for stretch goals)
- Performance Rating Distribution: % in each tier
- High Performer Retention: % of top 20% retained (Target: 95%+)
- Low Performer Management: % of bottom 10% exited or improved (Target: 80%+)
- Time-to-Proficiency: Days to full productivity (varies by role)
- Internal Mobility Rate: % moving to new role internally (Target: 15-20%)
Engagement
- Employee Engagement Score: Average survey rating (Target: 7.5/10 or 4.0/5.0)
- eNPS (Employee Net Promoter Score): % promoters - % detractors (Target: 30+)
- Survey Participation Rate: % completing survey (Target: 80%+)
- Manager Effectiveness Score: Direct report ratings (Target: 4.0/5.0+)
- Pulse Survey Trends: Real-time sentiment tracking
Development
- Training Completion Rate: % of assigned training completed (Target: 90%+)
- Training Hours per Employee: Annual hours (Target: 40+)
- Skill Gap Closure: % reduction in gaps after training (Target: 50%+)
- Internal Promotion Rate: % of positions filled internally (Target: 30-40%)
- High-Potential Identification: % of workforce (Target: 10-15%)
Diversity, Equity, Inclusion
- Workforce Diversity: % representation by demographics
- Leadership Diversity: % in leadership roles
- Pay Equity: Compensation parity (Target: <5% unexplained variance)
- Inclusion Score: Survey rating on feeling included (Target: 4.0/5.0+)
- Diverse Candidate Slate: % of finalist pools with diversity (Target: 50%+)
Productivity
- Revenue per Employee: Total revenue / headcount
- Profit per Employee: Total profit / headcount
- Absenteeism Rate: % of scheduled days missed (Target: <3%)
- Overtime Hours: Average per employee
- Span of Control: Direct reports per manager (Target: 5-10)
Data Requirements for Predictive Analytics
Data Quality
- Clean: Consistent data across all HR systems
- Standardized: Common formats and definitions
- Validated: Regular audits and quality checks
- Complete: Minimal missing values
Data Volume
- Minimum: 2 years of comprehensive employee data
- Sample Size: 200+ employees for statistical significance
- Variables: Multiple data points across different dimensions
- Outcomes: Historical results to train models (who left vs. stayed)
Data Sources
- HRIS: Demographics, tenure, job history, compensation
- Performance Management: Ratings, goal achievement, feedback
- Talent Acquisition: Hiring source, time-to-hire, interview scores
- Learning Management: Training completion, certifications
- Engagement Surveys: Satisfaction, manager effectiveness, culture fit
- Payroll: Compensation changes, bonuses, equity
- External Data: Market conditions, industry trends, competitor intelligence
Feature Engineering
- Compensation history and changes
- Training records and skill development
- Manager changes and relationship quality
- Peer feedback and collaboration patterns
- Promotion velocity and career progression
- Work-life balance indicators (hours, PTO usage)
- External factors (market conditions, commute distance)
Analytics Tools and Platforms
Specialized HR Analytics Platforms
- Visier: Comprehensive people analytics, predictive models, dashboards (17x more accurate than guesswork for exit risk)
- HRBench: Predictive analytics for 25+ HR metrics, visual forecasts
- isolved: All four analytics types, easy dashboards, enterprise-grade
- Crunchr: Workforce planning and scenario modeling
- ChartHop: Org planning and people analytics
Business Intelligence Platforms
- Microsoft Power BI: Data visualization, integrates with Microsoft ecosystem, predictive analytics
- Tableau: Advanced visualization, Salesforce Einstein AI integration
- Qlik Sense: Associative analytics, self-service BI
- Looker (Google): Cloud-native, real-time dashboards
All-in-One HRIS with Analytics
- Workday: Prism Analytics, predictive insights, benchmarking
- SAP SuccessFactors: Workforce Analytics, People Analytics
- Oracle HCM Cloud: Analytics Cloud, AI-powered insights
- BambooHR: Standard and custom reports, basic analytics
Machine Learning Algorithms
- Decision Trees: Classification (will employee leave?), easy to interpret
- Logistic Regression: Binary outcomes (promote/don't promote), probability scores
- Random Forest: High accuracy, handles large datasets
- Neural Networks: Complex pattern recognition, large data requirements
- Transformer Neural Networks: Time series analysis, sequential data
- Clustering (K-Means): Segmentation (employee personas, performance groups)
- Natural Language Processing (NLP): Text analysis (survey comments, exit interviews)
Implementation Roadmap
Phase 1: Descriptive Analytics (Months 1-3)
Goal: Build dashboards for key metrics
Actions:
- Identify critical metrics to track
- Ensure data quality and integration across systems
- Build dashboards in BI tool (Power BI, Tableau)
- Establish baseline measurements
- Create regular reporting cadence
Deliverables:
- HR metrics dashboard (turnover, headcount, time-to-hire, engagement)
- Monthly/quarterly reports
- Data governance and quality standards
Phase 2: Diagnostic Analytics (Months 4-6)
Goal: Understand root causes of trends
Actions:
- Conduct correlation analyses
- Segment data for deeper insights (by department, role, demographics)
- Investigate anomalies and trends
- Root cause analysis on key issues (turnover spikes, engagement drops)
Deliverables:
- Diagnostic reports on key issues
- Segmented analytics
- Insights and recommendations
Phase 3: Predictive Analytics (Months 7-12)
Goal: Forecast future outcomes
Actions:
- Select high-impact use case (e.g., turnover prediction)
- Build and validate predictive model
- Pilot with one department
- Refine model based on feedback
- Scale to organization
Deliverables:
- Turnover prediction model
- Risk scores for employees
- Pilot results and learnings
- Scaled predictive analytics
Phase 4: Prescriptive Analytics (Months 13-18)
Goal: Recommend actions
Actions:
- Develop recommendation engines
- Automate interventions where appropriate
- Integrate with HR workflows
- Measure impact of recommendations
Deliverables:
- Automated retention intervention recommendations
- Personalized development plan suggestions
- Integrated analytics in HR processes
Best Practices
1. Start with Business Questions
- Don't start with data looking for insights
- Start with problems and find data to answer them
- Examples: Why is turnover high? Which candidates will succeed? What drives engagement?
2. Ensure Data Quality
- Conduct data audit (completeness, accuracy, consistency)
- Standardize formats and definitions
- Establish data governance
- Regular quality checks and validation
3. Build Analytics Capability
- Hire or develop data analysts/scientists
- Train HR team on data literacy
- Partner with IT and data teams
- Consider external consultants for specialized projects
4. Start Small and Scale
- Begin with descriptive analytics and dashboards
- Progress to diagnostic, then predictive, then prescriptive
- Pilot predictive models with one use case
- Scale based on results and learnings
5. Ensure Privacy and Ethics
- Comply with GDPR, CCPA, and other regulations
- Anonymize data where appropriate
- Audit models for bias (gender, race, age)
- Transparent communication about data usage
- Never use analytics punitively
6. Integrate with Decision-Making
- Translate data into clear recommendations
- Embed analytics in existing HR processes
- Provide real-time dashboards for managers
- Enable self-service analytics
- Combine data with human judgment
7. Measure and Communicate Impact
- Track ROI of analytics initiatives
- Measure changes in key outcomes (turnover, engagement, productivity)
- Calculate cost savings and revenue impact
- Share results with leadership
- Use storytelling and visualization
Common Pitfalls to Avoid
- Analysis Paralysis: Don't wait for perfect data, start with what you have
- Ignoring Data Quality: Garbage in, garbage out - prioritize data quality
- Lack of Business Context: Analytics without business understanding is meaningless
- Overcomplicating: Start simple, add complexity as needed
- No Action: Insights without action are wasted - ensure recommendations are implemented
- Ignoring Privacy: Data privacy violations can be catastrophic
- Algorithmic Bias: Models can perpetuate existing biases - audit regularly
- Replacing Human Judgment: Analytics informs, doesn't replace human decision-making
Real-World Examples
HP - Project Insight:
- Predicted employee turnover with high accuracy
- Reduced turnover from 20% to <10% (2009-2011)
- Saved estimated $300M
- Enabled proactive manager interventions
Google - Prediction Engine:
- Analyzes job history, education, skills, personality to predict candidate success
- Improved hiring quality and reduced mis-hires
- Project Aristotle identified elements of effective teams
Best Buy - Engagement-Revenue Link:
- Linked 0.1% engagement increase to $100K revenue per store
- Measured engagement more frequently
- Implemented targeted interventions
Cisco - Workforce Planning:
- Predictive analytics for skills gap forecasting
- Proactive reskilling and hiring
- Reduced time to fill critical roles
American Express - Remote Work Transition:
- Predicted when employees would need IT support
- Proactive support deployment
- Smooth transition to remote work
Unilever + HireVue - Video Interview Analysis:
- Machine learning analysis of video interviews
- 90% faster hiring process
- £1M annual cost savings
Using the Reference Files
When to Read Each Reference
/references/people-analytics-frameworks.md — Read when developing overall people analytics strategy, understanding analytics maturity models, or designing analytics governance. Covers strategic frameworks, organizational readiness, and analytics program design.
/references/predictive-modeling.md — Read when building predictive models for turnover, hiring, performance, or other HR outcomes. Includes detailed modeling techniques, algorithms, validation methods, and implementation guidance.
/references/hr-metrics.md — Read when defining which metrics to track, building HR dashboards, or establishing baseline measurements. Comprehensive catalog of HR metrics across all functions with targets and calculation methods.
/references/analytics-tools.md — Read when selecting analytics platforms, evaluating BI tools, or building analytics infrastructure. Covers tool comparisons, implementation considerations, and technology stack recommendations.