HR Analytics & Workforce Intelligence

Part of the worlds-biggest-software-project initiative.

An AI-native, open-source platform unifying HRIS, payroll, recruitment and engagement data to power attrition modelling, pay equity analysis, hiring funnel intelligence and engagement correlation.

Workforce decisions — who to hire, how to pay, who is at risk of leaving — are among the highest-cost decisions an organisation makes, yet most HR departments still rely on retrospective headcount reports and annual engagement surveys. This project brings explainable predictive analytics, EEOC-aligned pay equity regression and longitudinal engagement-to-attrition correlation into a single platform aimed at HR teams that do not have an in-house data science function.

Why HR Analytics & Workforce Intelligence?

Voluntary attrition is expensive: replacement costs are estimated at 50–200% of an employee's annual salary, yet most HR teams still operate on retrospective headcount reports rather than individual flight-risk signals.
The EU Pay Transparency Directive enters full force in 2026, requiring employers with 100+ workers to report pay gaps publicly — elevating pay equity analytics from optional to compliance-critical.
Visier and Workday People Analytics target the enterprise: Visier is over-specified and expensive for mid-market and typically requires a six-month consultant-led deployment; Workday People Analytics is only valuable to organisations already running Workday HCM.
Specialist tools are siloed: Culture Amp focuses on engagement surveys but is not a full HR analytics platform; Orgnostic addresses pay equity and hiring funnel but has fewer pre-built predictive models and limited workforce planning.
Model explainability is a regulatory pressure point: attrition predictions trained on historical data can encode structural biases, and regulators and employees increasingly expect explanations for algorithmic HR decisions — a known gap in incumbent enterprise tools.

Key Features

Headcount, Attrition & Workforce Planning

Headcount dashboard tracking hires, departures, transfers and net change by department, tenure and date
Attrition rate analysis segmented by voluntary/involuntary, department, manager and tenure band
Individual attrition risk scoring drawing on tenure, performance trajectory, manager change and compensation relative to market
SHAP-based explainability for each individual flight-risk score, supporting HR review and GDPR-jurisdiction compliance
Workforce planning scenario modelling for headcount and cost projection across departments and fiscal periods

Pay Equity & Compliance

EEOC-aligned multiple regression analysis identifying unexplained pay gaps by gender, ethnicity and job family after controlling for legitimate differentiators
Confidence intervals and sample-size warnings for small-population sub-groups
EU Pay Transparency Directive reporting workflow with structured gap-analysis output formatted for regulatory submission
Drill-down dashboards from organisation level to specific job families and business units

Hiring Funnel Intelligence

Conversion-rate tracking at each recruitment stage by channel, role type, recruiter and demographic group
Sankey-style funnel visualisation showing candidate drop-off
ATS integration with Greenhouse, Lever and Ashby for automated funnel data capture
Predictive time-to-fill modelling for open requisitions based on historical patterns

Engagement Correlation

Ingestion of engagement survey data (Culture Amp, Glint) for longitudinal sentiment tracking
Correlation engine linking survey sentiment scores to subsequent 90-day attrition and performance outcomes
Manager effectiveness signals combining team-level engagement, attrition and performance data

Access, Privacy & Governance

Role-based access control with manager, HR business partner and CHRO views
GDPR-compliant data handling with configurable data residency
Column-level security for sensitive pay data in the analytical warehouse

AI-Native Advantage

Where incumbents bolt natural-language Q&A onto proprietary dashboards, this project treats AI as a first-class layer: automated weekly flight-risk digests highlighting employees whose behavioural signals have changed, natural-language people analytics queries ("which departments are at highest attrition risk this quarter and why?"), AI-assisted pay equity narrative commentary for HR and legal review, and predictive hiring funnel modelling. Every predictive output ships with explainable feature contributions so HR teams can defend decisions to regulators and employees.

Tech Stack & Deployment

The expected stack pairs Python (scikit-learn, XGBoost, SHAP) for attrition modelling with R (equaltestpp, lm) for EEOC-aligned pay equity regression, dbt as the transformation layer joining HR, payroll and engagement data into a unified people model, and Snowflake or BigQuery as the analytical warehouse with column-level security. Apache Airflow orchestrates monthly payroll and headcount snapshot jobs, and HashiCorp Vault manages secrets and encryption keys. Connectors target Workday, SAP SuccessFactors and BambooHR for HRIS, Greenhouse, Lever and Ashby for ATS, and ADP, Sage and Gusto for payroll. EU data residency is a deployment-mode requirement rather than an optional feature.

Market Context

The global workforce analytics market was valued at USD 2.37 billion in 2025 and is projected to reach USD 7.12 billion by 2034, driven by AI accessibility, rising hiring costs and the acceleration of mandatory pay transparency legislation (AIHR). Primary buyers are CHROs, HR business partners and people analytics leads at mid-market organisations underserved by Visier's enterprise pricing and unable to standardise on Workday People Analytics. The EU Pay Transparency Directive's 2026 enforcement deadline is a compliance forcing function across European markets and is influencing legislation elsewhere.

Project Status

This project is in the research and specification phase.
Contributions, feedback, and domain expertise are welcome.

Contributing

We welcome contributions from developers, domain experts, and potential users. See CONTRIBUTING.md for guidelines.

Important: All contributions must be your own original work or clearly attributed open-source material with a compatible licence. Copyright infringement and licence violations will not be tolerated and will result in immediate removal of the offending contribution. If you are unsure whether a piece of code, text, or other material is safe to contribute, open an issue and ask before submitting.

Licence

Licence to be determined. See discussion for context.