---
id: "action-mine-workforce-data"
type: "action-item"
source_timestamps: ["§ A Playbook to Customize Scheduling"]
tags: ["data-analytics", "root-cause-analysis"]
related: ["concept-lasso-regression-workforce", "framework-customized-scheduling-playbook", "prereq-workforce-management-systems", "prereq-advanced-analytical-capability"]
action: "Apply advanced analytics to existing workforce data to identify localized drivers of turnover."
outcome: "Identification of the specific scheduling dimensions (e.g., fatigue vs. fairness) causing churn in specific locations."
speakers: ["Santiago Gallino", "Borja Apaolaza"]
sources: ["tail1"]
sourceVaultSlug: "hbr-seg-tail1"
originDay: 1
articleStem: "hbr-tail-111-service-worker-churn"
sourceUrl: "https://hbr.org/2026/03/the-solution-to-service-worker-churn"
sourceTitle: "The Solution to Service-Worker Churn"
---
# Mine existing workforce data for local turnover drivers

**Action:** Apply advanced analytics to existing workforce data to identify localized drivers of turnover.

Instead of collecting new data, use the raw data your workforce management systems already capture — timestamps, shift patterns, approvals, absences — that is currently used only for payroll and compliance. Apply [[concept-lasso-regression-workforce|LASSO]]-style analytics to **segment by location, store format, and worker group** to identify the specific [[concept-scheduling-quality-dimensions|dimensions]] driving turnover in each context.

This is **Step 1** of the [[framework-customized-scheduling-playbook|playbook]]. It depends on [[prereq-workforce-management-systems|data-rich workforce management systems]] and [[prereq-advanced-analytical-capability|advanced analytical capability]].

**Expected outcome:** Identification of the specific scheduling dimensions (e.g., fatigue vs. fairness) causing churn in specific locations.