Talent analytics: 4 strategic questions
Step 1 / 3·Case study: Google Project Oxygen
0/3 steps done
Learning objectives — after this module you will:
- Grasp the 4 analytics problems: retention, readiness, capability gap, mobility
- Learn from the Google Project Oxygen case
- Distinguish descriptive – diagnostic – predictive analytics
Case study: Google Project Oxygen
See the power of letting data test beliefs.
In the late 2000s, Google's engineering culture held a common belief: managers are a redundant layer — good engineers don't need a boss. The People Analytics team decided to TEST IT WITH DATA instead of arguing: Project Oxygen analyzed thousands of performance reviews, surveys, and nominations, comparing the best managers' teams with the lowest.
The result overturned the belief: teams with good managers clearly outperformed on retention, satisfaction, and performance. More importantly, Google distilled the specific BEHAVIORS of great managers (initially 8, later expanded to 10): be a good coach, empower without micromanaging, care about members' success and well-being, give clear direction, help with career development... The #1 ranking was a surprise: BE A GOOD COACH — not being the most technically skilled.
Google turned the finding into action: periodic manager-feedback surveys along exactly these behaviors, manager training tied to each behavior, and honoring excellent managers. The transferable lesson: analytics creates value when it (1) starts from a contested business question, (2) distills into DOABLE behaviors, (3) closes the measure–teach–re-measure loop.
Key takeaway: Project Oxygen: data tests beliefs, distills specific behaviors, closes the measure–teach–re-measure loop.