Educational Guide
HR analytics is the practice of turning workforce data into better people decisions. This guide explains what it covers, how it differs from HR reporting, and how to get value in a week without a data scientist.
HR analytics is the practice of collecting workforce data, turning it into metrics, and using those metrics to make better people decisions. It replaces "I think attrition is up in sales" with "first-year attrition in sales rose from 9% to 17% over two quarters, and it tracks with the onboarding change in March."
Reporting tells you what happened: headcount is 512, turnover was 14%. Analytics tells you why, and what to do about it. The two are often confused — HR reporting vs HR analytics covers the difference in detail.
| Level | Question it answers | Example |
|---|---|---|
| Descriptive | What happened? | Turnover was 14% last quarter |
| Diagnostic | Why did it happen? | 60% of leavers were in their first year |
| Predictive | What is likely to happen? | Teams with low manager tenure show higher forecast attrition |
| Prescriptive | What should we do? | Move onboarding investment to the two highest-risk teams |
Most teams start at descriptive and work down as their data matures. You do not need predictive models to create value — a clear diagnostic view already changes decisions.
At a minimum, an employee master file with: employee ID, hire date, exit date (where applicable), department, location, job level, and — where lawful and appropriate — pay. Add recruitment, absence, training and cost data later to unlock more metrics. HRlytix reads Excel or CSV and maps the columns for you; see uploading data.
Start with three or four you will actually act on, not thirty.
HR Metric
Turnover Rate
All exits (voluntary and involuntary unless configured otherwise) divided by average headcount.
Exits During Period / Average Headcount × 100 · Average Headcount = (Beginning HC + Ending HC) / 2
Required data: employee_id, termination_date, employment_status
View metric definition →HR Metric
Retention Rate
Employees active at the beginning of the period who remain active at the end, as a percentage of the beginning population.
Employees Remaining at End / Employees at Beginning × 100
Required data: employee_id, employment_status, hire_date
View metric definition →The HR KPIs that drive decisions has a fuller shortlist with formulas.
Note
Be careful with "industry average" figures. A benchmark only means something if you know its source, year, industry, geography and how the population was defined. If you can't verify those, treat it as noise — your own trend over the last four to eight periods is usually a better yardstick.
The terms are used interchangeably. "People analytics" is more common when the work also covers engagement, performance and organisational-network data, not just core HR metrics.
No. Descriptive and diagnostic analytics need clean data and good questions, not modelling. Bring in specialists when you move to prediction.
One period gives a snapshot. Four or more periods let you see trends and comparisons, which is where most decisions come from.
Every metric in this guide is defined — with its formula and required fields — in the HR Metrics Dictionary.
Open the HR Metrics DictionaryData-driven HR means workforce data is the first thing you look at for a people decision, not the last. A decision framework, practical examples, and the privacy guardrails that keep it responsible.
Oct 12, 2026 · 7 min read
How to calculate employee turnover rate, the voluntary/involuntary split, why the denominator matters, and how to investigate a rise instead of guessing.
Oct 12, 2026 · 6 min read
How to calculate retention rate, why it is not just "100 minus turnover", how to read a retention curve, and the mistakes that make the number misleading.
Oct 12, 2026 · 6 min read
HRlytix calculates every metric in these guides from your own Excel or CSV data — no formulas required.
Try HRlytix free