What the policy is for
An absenteeism policy is not primarily a deterrent. Employees who are ill are ill whatever the handbook says, and employees who have disengaged are not reading it.
Its function is to make the organisation's response predictable. Without one, absence is handled by individual managers according to temperament, and the same record produces sympathy in one team and a warning in another. That inconsistency is what employees notice, what they discuss, and what turns an ordinary attendance issue into a grievance.
A policy that achieves consistency and nothing else has done its job.
What it has to define
Four definitions carry most of the weight.
| Clause | What it should establish | Where it goes wrong |
|---|---|---|
| What counts as unplanned absence | Absence not approved in advance, whatever the reason | Defining it by reason, so illness is excluded and the measure stops reflecting disruption |
| How and when to notify | Who to tell, by when, and by what means | A rule requiring a call to a named manager before shift start, with no alternative if they do not answer |
| What evidence is required, and when | The threshold at which a certificate is asked for | Requiring evidence for a single day, which costs more in goodwill than it recovers |
| What follows from repetition | The trigger points, and what happens at each | Automatic escalation with no room to distinguish a chronic condition from avoidance |
The notification clause deserves particular attention, because it is the one most often written by someone who has not recently been ill. A rule that requires a personal telephone call to one named individual before the shift starts fails whenever that individual does not pick up, and it fails hardest for the employee who is genuinely too unwell to make the call. Name a fallback and accept a message.
Triggers, and what they should trigger
Most policies set trigger points, and most set them to trigger the wrong thing. A trigger should start a conversation. Only a pattern established through those conversations should start a process.
- Return-to-work discussion after every absence. Short, routine and not disciplinary. This is the highest-value clause in the policy.
- A review conversation once absence passes a stated frequency in a rolling period, to establish whether there is an underlying cause.
- A formal process only where the review has happened, a cause has been established or ruled out, and the pattern continues.
Automatic progression from a score to a warning is the defect worth designing out. It treats a chronic condition and deliberate avoidance identically, which is unfair on its face and difficult to defend if challenged.
It is also worth stating explicitly that absence covered by a statutory entitlement does not count towards a trigger. Otherwise the policy quietly penalises the exercise of a right, which is the sort of clause that reads badly when read back.
What the policy should not try to do
Three things belong outside it.
It should not restate statutory leave entitlements. Those come from the applicable state shops and establishments enactment or the Occupational Safety, Health and Working Conditions Code, they vary by state and establishment, and a handbook that paraphrases them will eventually paraphrase them wrongly. Reference them and keep the numbers in one place.
It should not attempt to define acceptable reasons for absence. Any list will be incomplete, and the omissions read as judgements about whose circumstances count.
It should not carry the disciplinary procedure. Cross-reference it. A policy that contains a full disciplinary process is one nobody reads to the end.
What it should carry, and frequently does not, is a line on what happens to an employee's pay and leave balance when absence is unpaid, since that is the question they will actually ask.
Making it work in practice
A policy is only as consistent as the managers applying it, and the common failure is that they have never been told how.
- Give managers the return-to-work conversation as a short script rather than as an instruction. Most avoid it because they are unsure what to say.
- Show each manager their own team's absence data. Consistency is easier to ask for once the variation is visible.
- Record the conversation, not just the absence. A pattern with no record of anything ever having been raised is not a pattern anyone can act on later.
- Review trigger points against what actually happens. A threshold that fires on half the workforce is not a threshold.
Where a policy is being written for the first time, the useful sequence is to look at the current distribution first. Trigger points set without reference to the existing pattern tend to be set either where nobody reaches them or where everybody does.
Frequently asked questions
What should an absenteeism policy include?
A definition of unplanned absence based on whether it was approved in advance, a notification rule with a workable fallback, the point at which evidence is required, the trigger points for a conversation and for a formal process, and a line on the effect of unpaid absence on pay and leave balance.
Can an absenteeism policy override statutory leave?
No. Entitlements under the applicable state enactment or the Occupational Safety, Health and Working Conditions Code apply on their own terms, and a policy cannot reduce them. Absence covered by such an entitlement should be stated not to count towards trigger points.
Should the policy require a medical certificate?
For longer absences it is reasonable. Requiring one for a single day usually costs more in goodwill and administration than it recovers, and it pushes employees towards taking two days so the certificate is worth obtaining.
What is a return-to-work conversation?
A short discussion after each absence covering what happened and whether anything is needed. It is the most effective single clause in most policies, and it works by making absence noticed rather than by being punitive. It should not be run as a disciplinary meeting.
Should trigger points automatically lead to a warning?
No. Automatic escalation from a score to a warning treats a chronic condition the same as deliberate avoidance, which is unfair and hard to defend. A trigger should start a conversation; only an established pattern should start a process.
How Engage applies an absence policy
Engage records absence with its category and approval state, so unplanned absence is separable from approved leave without a manual reconciliation. Trigger points can be read against the actual distribution before they are set, and the conversation held after an absence is recorded against the employee alongside it, which is what makes a pattern defensible later.
See leave management in Engage