Most maintenance software is sold as a feature list. That is not how it arrives in a plant. It arrives as six people with different jobs, different screens, and very different days — and it works only if each of them gets what they need, when they need it, without being handed someone else’s complexity.
So here is one ordinary Tuesday at a mid-sized beverage filling plant: a filling line, a packaging line and a utilities area. Same plant, same shift, following the six maintenance roles the system uses — Supervisor, Technician, Mapper, Manager, Admin, and Viewer.
06:30 — The Supervisor opens the dashboard
Elif is the shift supervisor: coffee, portal, one screen. Seven cards across the top answer the only question that matters before the crew arrives: what is waiting?
Overdue Tasks counts work still open and past due as of today — it does not quietly reset when Elif changes the date range, so last week’s backlog cannot hide behind this morning’s view. Unassigned Tasks are active work with nobody’s name on it. Repair Requests are problems technicians reported while doing something else. Alerts are measurements above safe limits. Remarks, Notes, and NFC Skipped fill in the rest.
Then Elif distributes the day. The Assign Task page shows unassigned work as a tree on the left and technicians on the right; filter to the filling line, select the morning’s work, tick a name, assign. One technician per task is the normal setting — you can switch on assigning the same task to several people per plant, and when it is on, the task is duplicated to each of them rather than divided between them.
The calendar handles the rest of the week. A planned task can be dragged to another day; a finished one cannot be dragged at all, because the day a piece of completed work is recorded against is history rather than a plan.
What a manager cares about: the shift is distributed before the crew clocks in, and nothing waiting is invisible.
07:00 — The Technician’s phone
Jonas is a maintenance technician and will not see the portal all day. The phone was paired once, by scanning a QR code the Admin generated, and it has remembered the configuration ever since. The app keeps Jonas signed in, so it opens straight to the day’s work—which is why the phone has its own screen lock: that lock, not the app, is what keeps the work private when the phone lies on a toolbox.
The home screen is three tiles: what is to do, what is completed, and the preparation list. The task list underneath is grouped by due Date. Estimated hours for today’s open work sit at the top, answering the only planning question a technician has at 07:00.
Two checklists run before the first task of the day — once, not on every job. The safety checklist combines every safety procedure configured for the company into one list to tap through. The preparation checklist gathers the tools and products today’s tasks need, so the walk to the line happens once with the right grease in hand. On a machine that requires lock-out, a separate confirmation appears on that task point every time, not once a day.
The app is offline-first — not a fallback mode, the working assumption. Jonas can execute a whole shift in a steel-clad hall or a basement with no signal: forms, photos, repair requests, all of it queues on the device and uploads by itself on the way back. The app is also in Jonas’s own language; it ships in fourteen.
What a manager cares about: no signal is no excuse, and no one works from memory—more on the Mobile App page.
08:10 — At the machine
At the filling line, Jonas holds the phone against the machine’s NFC tag, and the right task point opens—no scrolling through a list of near-identical names, no chance of filling in the form for the pump next door. A barcode does the same job on machines where a printed label fits better. If more than one task is queued for that point, the app offers the choice.
The form is whatever this task type needs. Readings are typed against an expected range: where soft limits are set, a value at the edge highlights amber, anything outside the range highlights red, and an out-of-range figure has to be confirmed before the task will go through. That confirmed reading becomes an Alert on Elif’s dashboard.
A photo of the gauge is added to the task. A remark is chosen for the next shift. And when Jonas finds something genuinely wrong — a weeping seal on the carbonator — a repair request is raised from inside the task, with a photo, which the form requires. It lands in the Repair Requests card Elif looked at this morning.
One tag on the packaging line has been painted over. Jonas records an NFC skip with a reason. Skipping is only possible where the plant allows it; each skip is counted, and each one lands in a review queue where Elif can accept it or flag disagreement later — so an unreadable tag becomes a job for the label printer instead of a habit.
What a manager cares about: the record says which machine, at what time, with a picture — and the exceptions are visible rather than absent.
09:30 — The reading round
Some machines carry a counter — running hours on the compressors, cycles on the capper — and some work exists purely to go and read them. Those arrive as a reading round: one scheduled task listing the counters to read, grouped by machine, in the order the round is walked.
Every counter needs an answer before the round submits: a value, confirmation that the machine stood idle, or the reason it could not be read. Under each field is the last reading and how old it is so that you can judge a number against something. Service intervals can then be planned based on what a machine has actually done rather than the calendar.
What a manager cares about: counters behind usage-based servicing get read on a plan, and a missed round is visible. We wrote about this in detail in Service by Running Hours, Not by the Date.
11:00 — The Mapper at the desk
Priya is the maintenance planner — the Mapper role — and owns the structure everything else hangs from: production site, production line, section, machine, task point, task. It is the part of maintenance software for plants that nobody sees, and everybody depends on.
A second capper arrived last month. Rather than authoring it from nothing, Priya duplicates a task point from its twin on the other line; the copy carries the original’s flags across and is scheduled straight away. Shutdown-only and inactive tasks are deliberately left behind, and the portal says how many, because those are decisions someone made rather than defaults to propagate.
Schedules are set per line, machine, or section: daily, weekly on chosen weekdays, twice weekly, or monthly on an ordinal weekday — “last Friday of the month” means the actual last Friday. If a job genuinely needs several visits in one day, you can switch it on per plant, and each visit gets its own start time.
Priya writes the new NFC tag from the phone and sticks it on the capper. Products and tools attach to tasks, which makes Jonas’s preparation checklist correct tomorrow morning. A decommissioned machine is deactivated rather than deleted — it disappears from everyone’s day while its history stays intact.
What a manager cares about: a new machine is productive in an afternoon, not a project. See Asset Management.
13:00 — Shift change, and a pump that fails
A technician calls in sick. Elif filters to their name, selects their work — including what has already gone overdue — and reassigns it to the afternoon crew in one move rather than task by task.
Then a utilities pump fails, and the line has to come down. A shutdown is where two things meet: repairs that were deliberately deferred to the next stop, and tasks marked shutdown-only because they cannot be done on a running line. Both appear in the shutdown’s work list, scheduled inside the shutdown window, and that is what the technicians see on their phones.
Where a plant wants a second pair of eyes, approvals can be switched on. When they are — and when a task is flagged to need it — the submission waits for Elif: values, photos, remarks. Approve, and it closes. Reject, and it reopens in the technician’s own list with the reason attached, visible on the form when they open it again.
What a manager cares about: an unplanned stop becomes the moment the deferred backlog gets cleared. See Safety & Shutdown.
15:30 — When something slips
A lubrication task on the packaging line is still open well past its time. Where escalation is switched on, an overdue task climbs the chain by itself: the Supervisor first, then Managers and Admins.
Each person sets how it reaches them. Every notification lands in the in-app inbox; push and e-mail add extra reach, chosen per category, and quiet hours keep a phone silent overnight. An escalation asks to be acknowledged, and acknowledging it stops it climbing further — the alert has demonstrably reached a person.
What a manager cares about: nothing depends on someone remembering to check.
17:00 — The Manager’s view
Daniel manages maintenance across the plant and is not interested in individual tasks. Completion rate over time shows whether the trend is holding. Technician performance lists assigned tasks, completion status, average time to completion, and rejection rate—the basis for a coaching conversation rather than a guess. Line availability shows uptime and downtime per production line.
When something needs explaining, the Activity Log is an append-only record of who changed what and when, and Global Search finds any machine, task, or user from anywhere in the portal. A company with several plants sets scope once in the header switcher and the whole portal follows.
Sofia, the Admin, keeps the rest running in the background: users and roles, company settings, languages, master data, and the switches that turn optional parts on—approvals, assigning one task to several technicians, several visits a day, and the inventory module. None of those are on by default; each is a decision a plant makes.
What a manager cares about: the numbers come from the work itself, not from a spreadsheet somebody maintains. See Reporting.
Monday 07:00 — The weekly report
A roll-up of the week’s activity arrives by e-mail on Monday morning for Managers, Supervisors and Admins. Nobody assembled it.
This is also where the sixth role earns its place. A Viewer — a quality auditor, a visiting certifier, a member of the management team — sees the hierarchy, the tasks, the schedules and every standard report with export, and changes nothing. Read-only is the whole point: audit access without a second thought about what might get edited.
Plants that want it can also switch on the optional inventory module, where stock balances move as technicians record what they actually used on a task. It is an add-on rather than part of the standard product.
Who does what
| Role | Usually | Works in | Sees first each day |
|---|---|---|---|
| Technician | Field technician, fitter, electrician | Phone | Today’s tasks and the preparation list |
| Supervisor | Shift or maintenance supervisor | Both | Overdue, unassigned and repair requests |
| Mapper | Maintenance planner or engineer | Both | The asset hierarchy and what needs authoring |
| Manager | Maintenance or plant manager | Portal | Completion rate, performance, line availability |
| Admin | Systems or operations administrator | Portal | Users, settings and optional features |
| Viewer | Auditor, quality, management | Portal | Reports and history — read-only |
Nothing in that day is dramatic, which is the point. The value of maintenance software for plants isn’t in any single feature — it is that six people with six different jobs each get a screen that fits their job, and handovers between them stop depending on a radio call or a clipboard someone left on a desk.
If you recognized your own plant in any of those scenes, the quickest way to judge whether the maintenance roles map onto your team is to walk one of your own lines through it. Book a demo and bring your most awkward shift — or start with the full feature overview.