Tickets expire on a schedule you already know. Planning against that calendar instead of discovering it at the gate changes what your roster can promise.

The most predictable surprise in the business

There is a category of event in workforce management that arrives with total predictability and is treated, every single time, as a surprise. The high risk work licence that expires mid-swing. The medical that lapses the week before a shutdown. The confined space ticket that quietly died in March and got discovered in July, at the gate, in front of everyone.

None of these are surprises in any honest sense of the word. The expiry date was printed on the card the day it was issued. It sat in a spreadsheet, or a system, or a photocopied wallet card in a personnel file, for one to five years, counting down in plain sight. An asteroid strike is a surprise. A ticket expiry is an appointment you chose not to put in the diary.

And yet across the industry, expiries are managed as if they were audit material rather than planning material. They surface in three places: a compliance report that looks backwards at how many are already expired, an audit finding that scolds everyone for the same thing with a reference number attached, and the gate, where the discovery is maximally expensive and maximally public. All three are downstream of the failure. All three are reactions. None of them are planning.

The reframe this article argues for is simple to say and surprisingly deep in its consequences: an expiry date is a piece of forward-looking roster data, exactly like a flight booking or a leave request, and it should live in the planning process, not the checking process.

What a compliance report cannot do

Most organisations of any size have some version of a compliance report. It typically says things like: 96 per cent of the workforce is fully compliant, 23 tickets expired, 41 expiring in the next 30 days. It gets produced monthly, circulated, and nodded at.

Notice what that report is. It is a photograph of the past. The 23 expired tickets are already failures, already people who either worked non-compliant or got stood down. The 41 expiring soon are a list without context, because the report does not know which of those 41 people are rostered onto a critical shutdown in week five and which are on leave for a month anyway. A ticket expiring next month for someone who is not required next month is a non-event. The same expiry for the only HV-qualified electrician on a shutdown crew is a schedule risk that deserves a manager's attention today.

The compliance report cannot tell those two cases apart because it lives in the compliance silo, disconnected from the roster. That is the structural problem. Compliance data without roster context is trivia. Roster data without compliance context is fiction. The value is entirely in the join.

The join: expiry meets roster

Put the two calendars on top of each other and a new kind of question becomes answerable, which is the whole point:

Who on next month's swing will have anything expired on the day they fly, not on the day the report was run? Which crews lose a required competency mid-swing, meaning the person gets on site fine and becomes non-compliant on day six of a fourteen-day roster? Which upcoming shutdowns depend on a competency held by exactly one rostered person whose ticket lapses within the shutdown window? What does the refresher training load look like for the next quarter if we book everything at the optimal time instead of in a panic?

None of these are exotic analytics. Every one of them is a date comparison between two lists your organisation already possesses. The reason they feel out of reach in most businesses is not technical sophistication, it is that the two lists live in different systems, owned by different teams, reconciled by nobody.

The mid-swing expiry deserves special mention because it is the one the gate cannot catch. Access checks happen at entry. A worker whose ticket is valid on arrival day badges through green, then goes non-compliant on Wednesday of week two while standing on your site. Nothing flags it. If something goes wrong on Thursday, the incident investigation will find it in about four minutes, and the paper trail will show your systems had the expiry date all along. That is an uncomfortable document to be on the wrong side of. Only a forward-looking, roster-joined view catches this case, which alone justifies building one.

What changes when expiries drive planning

Treating expiries as planning input changes behaviour in a chain, and each link saves real money.

Training gets scheduled into rosters instead of around them. The refresher stops being an emergency booked at whatever provider has a seat this week, at premium prices, on days that hurt the project. It becomes a planned activity slotted into R&R periods or low-demand windows, months ahead, at normal prices. The same course, the same person, radically different cost and disruption depending purely on when the decision was made.

The roster stops writing cheques the workforce cannot cash. When planners can see competency validity across the roster horizon, they stop assigning people to swings they will not be eligible to complete. The promise a roster makes to a project, these people, this window, these capabilities, becomes a promise backed by data rather than optimism.

Renewal becomes a pipeline, not a fire drill. With a rolling 90-day forward view, someone owns a steady, boring queue: book the medical, schedule the VOC, chase the licence renewal. Boring is the goal. Every expiry handled in the pipeline is a gate incident that never happened, a flight that never got changed, an awkward client conversation that never occurred. Prevention has no highlight reel, which is why it needs a process rather than heroics.

Audits become a formality. Here is the pleasant irony. Organisations that treat expiries as audit material tend to fail audits, because the audit finds the backlog. Organisations that treat expiries as planning material breeze through audits, because by the time the auditor arrives there is nothing expired to find and a documented pipeline demonstrating control. The best audit preparation is not preparing for audits. It is running a process where the audit has nothing to catch.

Getting there is smaller than it looks

The temptation is to declare this a systems project and wait for a platform. Resist that. The minimum viable version is embarrassingly modest.

Start with one join: next swing's crew list against the expiry register, matched by name, flagging anything that lapses within the swing window plus a fortnight. That is one afternoon in a spreadsheet if it has to be. It will be ugly. It will also, almost certainly, find problems on its very first run, and every one of them is a gate turnaround you just bought back for the cost of a VLOOKUP.

Then make it a rhythm. The forward expiry review becomes a standing ten-minute item in the weekly planning meeting, alongside flights and beds, with a named owner and a rule: nothing inside the window without a booked fix. That rhythm matters more than the tooling. Systems make it scalable and less fragile, and eventually you will want the join automated, with real identity matching instead of names, and expiry data flowing from source instead of being retyped. But the rhythm is the reform. The software just makes the rhythm cheap.

One more thing worth writing down: assign the expiry to a person, not a team. "Compliance monitors expiries" is how expiries end up on a report. "Kylie owns the 90-day pipeline and reports exceptions weekly" is how they end up handled. Every ticking date should have exactly one name next to it.

Appointments, not asteroids

The industry loses an extraordinary amount of money and goodwill to events it had complete advance knowledge of. Expiries are the purest example: every single one is announced years in advance, in writing, with a date. The only question any organisation gets to decide is where it will meet that date. In a planning meeting three months early, calmly, for the price of a course booking. Or at the gate at 5:30am, expensively, in front of the client.

The date does not move. The meeting place does. Choose the meeting room.