90-Day Turnover

90-Day Turnover: Percentage of new hires who leave within 90 days.

Formula

90-Day Turnover = Departures within 90 days / New hires × 100

Computable definition

The canonical formula in the GASP formula grammar. The MCP server compiles it to SQL over the datasets listed below.

count_distinct(field.employee_id where field.termination_date is not null and field.termination_date - field.hire_date <= 90) / count_distinct(field.employee_id where field.hire_date >= period_start and field.hire_date <= period_end)

Unit: percent (the formula yields an unscaled ratio)

Breakdowns

by_hire_source: one row per hire_source value

Notes: Cohort is new hires in the period. 90 days approximated as 3 months of tenure. by_hire_source returns one row per hiring source, which shows the channels that bring hires who leave inside 90 days.

Data requirements

Grain: Monthly aggregate · Frequency: Monthly · Aggregation: aggregate (month) · Staged: no · Main source: HRIS.

Field	Dataset	Type
`employee_id`	employee	string
`hire_date`	employee	date
`snapshot_date`	employee	date
`termination_date`	employee	date

Example SQL (Postgres)

SELECT (CAST((SELECT COUNT(DISTINCT employee_id) FROM employee WHERE snapshot_date = :period_end AND (termination_date IS NOT NULL AND (termination_date - hire_date) <= 90)) AS DOUBLE PRECISION) / NULLIF((SELECT COUNT(DISTINCT employee_id) FROM employee WHERE snapshot_date = :period_end AND (hire_date >= :period_start AND hire_date <= :period_end)), 0)) AS day_90_turnover

From an agent: with the GASP MCP server, call lookup_metric({ "name": "90-Day Turnover" }) or generate_query({ "metric": "90-Day Turnover", "adapter": "<your adapter JSON>" }).

Benchmarks

Below 5%: Good onboarding and hiring
5-10%: Average
Above 10%: Hiring or onboarding issues

What It Tells You

Hiring quality and onboarding effectiveness.

← Back to People metrics