Short answer
Employee turnover describes departures from a defined workforce during a stated period. A rise in that measure is an outcome to investigate. Once you have calculated a comparable rate, locate the pattern, check the working conditions around it and choose one proportionate action. Separate a broad rate from voluntary exits, planned changes, tenure and role context. Listen to current and departing employees without assuming any comment proves cause. Review whether the action changed the condition it was meant to address.
Start with a decision, not a generic retention programme
The turnover-rate guide explains the calculation. This guide begins after a leader sees a pattern and needs to decide what to investigate or change. The relevant question might be: should we alter the first 90 days for a role, change a shift handover, clarify progression, review a manager-support practice or accept a planned workforce change?
The CIPD workforce-planning factsheet treats workforce planning as matching changing organisational needs with people strategy. That is why total turnover alone is an incomplete starting point. A change in a seasonal population, planned restructuring or internal redeployment may require a different response from early exits in a role the organisation needs to keep staffed.
Use this turnover decision sheet
| Step | Question | Output |
|---|---|---|
| Define | What population, leaver rule and period are we looking at? | A reproducible turnover view |
| Locate | Where does the change appear by role, tenure, site, shift or time? | A small number of reviewable segments |
| Check context | What changed in the work, staffing, management, pay practice, access or progression? | Competing explanations to test |
| Listen | Who can describe the work before and after the pattern? | Contextual accounts, not an automatic conclusion |
| Act | What is one feasible change with an owner? | A bounded action and review date |
| Review | What evidence would indicate the condition changed? | A follow-up decision, not a promised retention result |
Keep departure information in proportion
The BLS distinguishes quits, layoffs and discharges, and other separations in its JOLTS definitions. Your HRIS categories may differ, but the principle applies: state the categories before treating a total as a retention problem.
Exit conversations can identify themes worth testing. They cannot tell you what every current employee experiences, and a current employee account does not prove that a given departure would have been avoided. Use both carefully. The CIPD's employee voice factsheet is useful context for choosing ways to hear employee experience; an organisation still needs to define who will review the material and respond.
A fictional diagnostic
This example is fictional. A regional logistics employer sees more voluntary exits among night-shift coordinators in their first six months. The rate calculation is checked first: the sites use the same employee population and leaver rule. The pattern remains, but only at two sites.
The review does not label the cause "poor engagement." It compares rota stability, training access, manager coverage, tool access and employee accounts. The accounts point to a recurring handover problem: new coordinators are responsible for exceptions before they can see the previous shift's unresolved work. At the other sites, experienced coordinators use a short handover record and a named escalation route.
The owner tests a shared handover routine at one affected site, gives the shift lead time to review it and asks new coordinators whether the unresolved-work context is clearer. The next review checks adoption, the reported handover condition and turnover over a comparable period. It does not claim the change caused every future employee to stay.
Choose actions by the condition, not by the metric
| What the review finds | A possible action to test | What not to assume |
|---|---|---|
| Role promise differs from daily work | Clarify hiring and onboarding materials with the operational owner | A rewritten description alone will resolve exits |
| New hires lack access, context or support | Improve the first-work-moment handover and manager check-in | Training completion proves readiness |
| Workload or rota design creates recurring friction | Review staffing, workload allocation or decision rights | Employees need only more resilience training |
| Progression is unclear | Publish the relevant criteria and schedule a career conversation | Interest alone proves a person is ready to move |
| A useful practice exists in another team | Test a local adaptation with the people doing the work | The practice transfers unchanged across all sites |
Where Lontra may fit
Lontra can help collect structured employee context for one bounded question, such as what makes a new-shift handover workable. It should complement HRIS records, manager judgement and the organisation's own processes. It does not predict who will leave, decide employment outcomes or prove a retention intervention worked. A trial is one campaign with up to 30 invitations over 60 days, without a card; managers receive briefs rather than raw conversations.
Explore employee conversations with Lontra and start with the operational question your team needs to answer.
Sources
Frequently asked questions
- What should HR do after turnover rises?
- Check definitions and where exits cluster, then test a focused explanation with relevant people and operational evidence. Choose an action with an owner and review date rather than applying one broad retention programme.
- Can exit interviews explain turnover?
- They can provide useful accounts after departure, but they are one input. Compare them with current employee voice, role and operational context without treating any one account as proof of a general cause.
- Does low turnover always mean healthy retention?
- No. A headline rate does not show whether workforce movement is voluntary, constrained, concentrated in critical roles or balanced with internal mobility and workforce needs.