Safety data only becomes useful when the pieces sit together. A hazard reported in March, an incident in the same area in June and a run of failed inspections on the same equipment are one story, but in most companies they live in three separate files owned by three different people, and nobody ever reads them side by side. The HSE module in Nexora holds all three. Observations, incidents and inspections share the same location structure, the same branches, the same evidence handling and the same reporting, so the pattern is visible without anyone having to go looking for it. This page covers all three in turn.
Observation cards go in a box on a wall and get emptied on a Friday, if somebody remembers. Incidents get written up in a document a day or two later, when the detail has already softened, and filed in a folder. Inspections get ticked on paper and typed up a fortnight afterwards from handwriting nobody can read.
Each of those is a familiar annoyance on its own. Together they are the real problem, because they mean no pattern can ever surface. The company genuinely does the work. It runs the inspections, it investigates the incidents, people do report hazards. What it cannot do is connect any of it, so every event is treated as the first time it has ever happened.
Then the auditor arrives and asks for the safety records for the year, and the safety manager spends two days assembling from three sources something that should have taken a click.
Observation reporting is the earliest warning a company gets. A loose handrail, a blocked exit, a spill nobody cleaned, a near miss that made someone’s stomach drop. Act on those and the serious incidents thin out. Make reporting slow or awkward and people stop bothering, and the first sign of trouble becomes an ambulance.
This part of the module is built to be fast at the one moment that matters, which is when someone is standing in front of the problem holding a phone. They take a photo, pick the location, describe what they saw, and it is in the system with a reference number before they have walked away. It runs in the phone browser with no app to install, which matters when a good share of the people on your site are contractors on their own devices.
Speed here is not a convenience feature, it is the whole product. Every extra field and every extra tap costs you reports, and the reports you lose are exactly the ones taken in a hurry by someone who nearly did not bother.
What it holds. Every observation carries what was seen in the reporter’s own words, photographs, the branch, and a precise location as area, then location, then sub location. That precision is what makes the data worth having. Three observations in the same sub location in a month is a story. Three observations described as “by the tanks” is a shrug.
Most incident systems have two states, open and closed. That is not an investigation, it is a filing cabinet with a lid. Nexora holds the whole chain instead. An incident is reported, an investigator is assigned, the investigation runs, root cause analysis happens as its own recorded stage, the findings are reviewed, and only then does it close.
Each stage is recorded with its timing, so a supervisor can see where every open investigation has got to and an auditor can see that the process was genuinely followed rather than summarised afterwards.
Severity that stays consistent. Six levels, applied the same way every time: fatality, serious injury, minor injury, property damage, near miss, no injury. Consistency is what turns a year of incidents into something you can reason about. It also keeps property damage and near misses in the same system as the injuries, which matters, because those are the cheap lessons. A near miss is an incident that told you what was coming and did not charge you for it.
The safety number, calculated properly. The lost time injury frequency rate is calculated from the incident records and from man hours entered per branch per month. Both halves live in the system, so the rate is derived rather than declared by somebody’s spreadsheet.
One honest limitation, because it matters more than a feature list. If a month of man hours has not been entered for a branch, the rate for that branch is calculated on incomplete hours and will quietly flatter you. Make man hours entry part of your monthly routine. We would rather say that on a website than have you find it out in a board meeting.
Inspections are where safety systems either become real or become paperwork. The task is simple: go to the thing, check the list, record what you found. What makes it hard is everything around it. Finding the right checklist. Knowing which piece of equipment you are actually standing in front of. Writing results on a clipboard and typing them in later, badly.
Here the inspector scans the QR tag on the equipment, the correct checklist opens against that exact asset, they work through it on the phone, and it is scored and filed before they walk away.
The QR tag solves the identification problem. Every asset in Nexora carries its own QR code. The inspector opens the scanner inside Nexora, points the phone at the tag, and the system knows exactly which unit it is, then opens the checklist belonging to that asset’s category.
This quietly fixes the biggest weakness in most inspection records. On paper, an inspector writes “extinguisher, workshop, checked”. Which extinguisher? There are eleven in the workshop. Was it the one that failed last quarter? Nobody knows, because the record points at a room rather than an object. Here every result is bound to a specific unit, which over time gives you a per unit history: this extinguisher has failed twice in eighteen months, that pump never has. Now equipment spending is an evidence based decision.
Scanning happens in the phone browser using the camera. There is no separate scanner app, which matters when the people running inspections change from month to month.
Templates that stay consistent. Checklist templates are set per inspection type by your own administrators, so everyone inspecting the same kind of thing asks the same questions in the same order. Templates attach to asset categories, so new equipment inherits the right checklist without anyone remembering to link it. Add twenty new extinguishers and all twenty are ready to inspect properly the same day.
Scored as it runs, with evidence where it counts. Compliance is calculated as the inspection happens, so when the inspector submits, the result is already a percentage recorded against the asset and visible on the dashboard. Nobody counts ticks and nobody types anything up. Any item can carry a photograph, and in practice the failed items are the ones that get photographed. A written note saying “guard damaged” is an opinion. A photo of the damaged guard is a fact, and it stays on the record permanently.
Inspections work the way they do because of a shared equipment register sitting under the module. Every asset has a code, a category and its own QR code, and the same record is used by the maintenance side of the platform. Deleting an asset is a soft delete, so history is never destroyed, because the accumulated history is the whole reason the register exists. That shared register is why an asset page can show its inspection history and its maintenance history in one view, which is the view that answers whether to repair something for the fourth time or replace it.
Area, then location, then sub location, with a branch on top. Set up once by your administrators and used identically by observations, incidents and inspections. It is the plumbing that makes the pattern visible, because three records describing the same place in three different ways will never line up.
