Module 1/4 22 min
People analytics and the role of AI
Step 1 / 4·The four levels of analytics and where AI fits
0/4 steps done
Learning objectives — after this module you will:
- Distinguish the four levels of people analytics
- Know the core HR metrics
- Understand why correlation is not causation
Learn ~6 min
The four levels of analytics and where AI fits
Have a frame for what 'analytics' means.
The four levels of people analytics
| Level | Answers the question | Example |
|---|---|---|
| Descriptive | What happened? | Last quarter's attrition was 8% |
| Diagnostic | Why did it happen? | Attrition concentrated in Dept A, the 1–2 year tenure group |
| Predictive | What might happen? | This group is at higher risk of leaving without intervention |
| Prescriptive | What should we do? | Prioritize retention, a development path for the at-risk group |
Where AI helps
AI lets you QUERY data in natural language ('which department has the most attrition and why?'), summarize findings, and suggest hypotheses. TTD's dashboards (Headcount & Attrition, Diversity, Engagement) already provide the descriptive and diagnostic parts; AI helps you interpret and tell the story from there.
Key takeaway: Four levels: descriptive → diagnostic → predictive → prescriptive; AI helps query, summarize, and interpret data.