Skip to content

Workforce & attendance

Who is on the desk right now, and who actually is?

A 24/7 assistance desk is a rota problem before it is a software problem. Four questions the platform previously could not answer at all.

Scheduled jobs
8, running per company
Finest cadence
Every minute
Roster states
Draft → Published → Locked
Handover
Against the real caseload

The problem

The rota lives in a spreadsheet, the duty doctor is whoever answered last time, leave is approved over chat, and when the coordinator holding six live cases goes home ill at 01:00 nobody finds out until a hospital calls.

What it does

Workforce & attendance, in practice.

  • Rosters that get published, locked and rolled forward

    Shift templates and rotation patterns generate a roster; it moves draft → published → locked, and rolls forward on a schedule every night rather than when somebody remembers.

  • Shifts, swaps and give-aways

    Assignments are accepted or released. Swaps and give-aways are validated when they are applied, not when they were proposed — the rota has moved since.

  • Attendance with corrections

    Clock in and out, breaks, overtime approval, and a correction path, because a desk that cannot correct a clock-in gets a desk that stops using the clock.

  • Leave, with displacement

    Request and decide, and the roster is told — approved leave displaces the shifts it lands on rather than sitting beside them in a different system.

  • On-call rotas and callouts

    Who is the duty doctor at this instant, including mid-period handover, and a record of the callout.

  • Coverage and a wallboard

    Coverage is checked on a schedule, not on request. The wallboard is the screen a supervisor leaves open.

  • Handover with orphan detection

    At the end of a shift, the outgoing desk hands over against the caseload it was actually holding — and the platform names what nobody picked up.

What this does not do yet: Push notifications and payroll export are not built. When a coordinator leaves mid-shift the platform raises the coverage gap and the handover snapshot; reassigning their live cases is still done by hand.

Rules that are not obvious

Why it behaves this way.

Each of these is a decision with a cost behind it. They are the difference between software that records what happened and software you can defend in an audit.

  • Two thirds of a rota's value is in noticing, and noticing needs a clock.

    Eight scheduled jobs run whether or not anybody opens the screen: shift transitions and on-call transitions every minute, attendance auto-close and swap expiry every five, coverage every fifteen, roster roll-forward and leave every night.

  • A run is recorded against the period it was for, not the moment it fired.

    On the night the clocks go back an hourly job sees 02:00 twice — and both ticks land in the same bucket, so the double-fire costs nothing. On a spring-forward morning a job set for an hour that does not exist runs at the first hour that does, bucketed as the hour it was meant to run at, instead of silently skipping a day.

  • Catch-up is declared per job, not guessed.

    Three thousand missed minutes of 'has anybody failed to clock in' is one sweep, because the answer comes from current state either way. Three missed nights of rota generation is three nights — collapse it and two days of the desk are simply absent from the schedule while the ledger records a success.

  • It is not payroll and not an HRMS.

    The exclusions are written down so nobody adds an hourly-rate column in six months and calls it an oversight.

See it against one of your own files.

The fastest way to judge CareTrack is to walk a real case through it — intake, guarantee, charges, adjudication, settlement, statement. Bring one and we will do exactly that.

Or email contact@sthiramtechnologies.com