An employee offboarding workflow: access removed, equipment back, nothing missed

SynaptaGrid8 min read

#workflow#offboarding

In short

An employee offboarding workflow is the set of steps between a confirmed leaving date and the last loose end: agreeing what is removed and returned, removing access, collecting equipment, and checking that nothing was missed. Most offboarding problems are hand-off problems. IT is told late, a laptop is never collected, or an account stays open because nobody owned the check. A good workflow names one owner per step, agrees the list before anything is removed, keeps a record of the case, and ends with a check that someone actually looks at. We built a small version in our own test workspace and ran it from start to finish, including a resignation that is withdrawn, and we describe what it produced and what it does not do.

What an offboarding workflow is

Offboarding is everything that has to happen so a person can leave cleanly: HR confirms the leaving date, the manager agrees what access is removed and what equipment comes back, IT does the removal and collects the equipment, and someone checks nothing was left open. A workflow writes those steps down, gives each to a person or a team, and keeps track of where each leaver stands.

Without one, offboarding runs on email and memory. It works until two people leave in the same week, or the one person who knows which systems the leaver used is on leave.

The version we built

Four steps where a person decides or confirms, two automatic steps that create and close a case record, one branch for a resignation that is withdrawn, and a closing step that records the result.

The steps to include, and who owns each

Confirm the leaving date
HR confirms the notice is accepted and the last day is fixed. Everything after depends on this, and it is also where a resignation that is withdrawn should end the process.
Open a case record
Once the date is confirmed, a record is created for the leaver with a status and a flag for equipment returned. It gives everyone one place to look, and it is created by the workflow rather than by someone remembering.
Agree the list first
The manager approves what is removed and what comes back, and who takes over the work, before IT acts. Agreeing afterwards means removing the wrong thing or missing something.
Remove access and collect equipment
IT removes the accounts and access on the agreed list and collects the laptop and badge. One owner, with the agreed list in front of them.
Hand over the work
Someone takes over the leaver’s files, tasks and customers. Add it as its own step if it is often forgotten.
Check nothing was missed
HR confirms access is removed, equipment is back and final pay is handled. This is the step most processes lack, and the one that finds the open account a year later.
Close the record and record the result
The case record is updated to closed with the equipment marked returned, and the workflow ends with a written result, so you can see how many offboardings finished and how many stopped.

Where approvals belong

Put an approval wherever a mistake is hard to undo. In offboarding that is the list of what is removed and returned: removing the wrong access can stop someone’s colleagues working, and missing one leaves a door open. Confirming the date, doing the removal and checking at the end are tasks, not approvals, so a person marks them done rather than approving them.

Decide who approves by role, such as the leaver’s manager, rather than by name. People change jobs, and a process that names an individual stops working the day they move.

The old way and the new way

The table compares offboarding run on email with offboarding run as a workflow.

Who knows the leaving date
The old way: Whoever was copied on the email
The new way: Every step reads it from the same place
What gets removed
The old way: Whatever IT remembers
The new way: A list the manager agreed, recorded at a named step
Where a leaver stands
The old way: Ask around
The new way: Open the run and see the current step, or open the case record and read its status
The record of the case
The old way: A spreadsheet someone updates when they remember
The new way: A record the workflow creates and then closes
A resignation that is withdrawn
The old way: Someone remembers to cancel the access removal
The new way: The process ends at the first step and no case record is created
Judgement
The old way: Still the manager’s and IT’s
The new way: Still theirs: what the leaver had, what is safe to remove

How we decided

We built the version in the diagram in our own test workspace in SynaptaGrid Automation, with a small record type in SynaptaGrid Data for the case, and ran it in November. In the full run we confirmed the leaving date and took every step to the end. The workflow created a case record with the status Offboarding started and the laptop not returned. At the end it updated that same record to Offboarding closed with the laptop marked returned, and the run finished with all steps completed and the result recorded. We checked the record in Data afterwards and it held the closed values. In the other run we chose the withdrawal at the first step: the run went straight to the closing step that records the withdrawal, finished in two steps, and no case record was created.

Three things went wrong first, and each is worth knowing. We wrote the branch conditions against the wrong field, so when the first step finished none of its branches matched and the run stopped as failed instead of waiting. The workspace has several Data apps, and the first run stopped with a message that it could not tell which one to write to; once we started the run against one app, the record was created. And the closing step could not find the record the opening step had made until we pointed it at the right place in that step’s result. In each case the run said what was missing rather than carrying on, but a test of every branch and every automatic step before anyone relies on it is not optional.

We chose agree-before-removal, a case record and a check at the end because those are the places the usual process fails. The choice of steps is ours and a judgement; it is not a standard.

Five limits, so this guide does not claim more than the runs showed. All the human steps were assigned to one test account, so the runs show the sequence and the branching, not hand-offs between different people. We did not connect the workflow to an HR, identity or equipment system, so nothing was removed from a real account: the only automatic actions wrote to a record in our own test data. We also set the workflow to send an in-app notification to that test account when the case closed, and no notification arrived in our test, so this guide does not describe notifications as working. We did not test reminders when a step waits too long. And the template carries one test account as the assignee and the actions tied to our test record type, so to reuse it you would change each step to a role and point the actions at your own record type, which we did not do.

Standards and accepted practice behind this

These are the practices the guide rests on. We align with them. This page does not claim a certification.

Least privilege
A person should hold only the access the work needs, so removing it when the work ends is part of keeping to that rule.
Source: NIST glossary: least privilege (opens in a new tab)
Segregation of duties
The person who agrees what is removed is not the person who removes it, and neither is the person who checks it at the end.
Source: NIST glossary: separation of duty (opens in a new tab)
Timely removal of access
Common security frameworks and audits expect access to be removed when someone leaves. A recorded step and a closed case record are the evidence an auditor asks for.
BPMN
The standard notation for drawing a process. The diagram above is a plain flowchart, and a BPMN diagram of the same steps would show the same decisions.
Source: OMG: BPMN 2.0 specification (opens in a new tab)

Where SynaptaGrid fits

SynaptaGrid Automation lets you write the steps as a workflow, assign each decision to a person, create and update records in SynaptaGrid Data as part of the run, and see where every run stands. You can start from the version above, change the steps to match how your company offboards, and keep the result as a template for the next leaver.

Check the limits above before you rely on it: whether each step is assigned to a role rather than a person, whether anything needs to reach your HR, identity or equipment systems, whether you need notifications, and who should be reminded when a step waits.

Frequently asked questions

What should an offboarding checklist include?
Confirming the leaving date, agreeing what access is removed and what equipment returns, removing access, collecting equipment, handing over the work, checking nothing was missed, and recording the result.
Who should approve what is removed from a leaver?
Usually the leaver’s manager, decided by role rather than by name, before IT removes anything.
How do I handle a resignation that is withdrawn?
End the process at the first step. Anything removed after that has to be put back, and a branch at the start prevents it.
Can a workflow keep the offboarding record for me?
Yes. In our test run the workflow created a case record when the date was confirmed and closed it at the end. You need a record type for the case and each action pointed at it.
Can I reuse an offboarding workflow for every leaver?
Yes, if you save it as a template and assign each step to a role. A template that names one person has to be edited first.
Share

SynaptaGrid

Build your offboarding steps

Start with the steps above in a test workspace and run one leaver through before you use it for real.