Site Management

Site Management is the on-the-ground side of CortexPM — who was on site, what happened, what’s defective, and who’s been briefed. Most of it is captured on the mobile app by the people actually standing on the slab, and reviewed in the web app by whoever’s running the job.

The point of all of it is the record. A diary backed by real sign-ins, a defect with a time-stamped photo and a named assignee, a toolbox talk with signatures against it — that’s what you reach for months later when someone disputes a delay or a rectification.

The Onsite Dashboard

Site Management opens on the Onsite Dashboard — who is on site, what got recorded, and what is still open.

Two tabs. Overview gives you the whole business at a glance:

• Site presence today — headcount by hour, against your typical weekday
• Currently on site — click through to the sign-in records
• Diary compliance — the last 14 working days. A day only counts as owed if the program booked work or somebody actually signed in, so a shutdown week doesn’t show as failure.
• Open defects — by trade, with how many are past their due date
• Toolbox talks — sign-off progress on active talks

An attention banner sits above the cards for the things you’d want pushed at you: the most recent site with no diary, and defects past due date.

Sites gives you one row per job — on-site count against expected, who’s absent, diary state, open and overdue defects, talk sign-off, whether the site is geofenced, and a red/amber/green dot. Click a row to open the job.

Across the top: Sign-in records, Diaries, Defects, Toolbox talks, Timesheets, Crew, Site boundary, and a Quick actions menu for New site diary, Log a defect and New toolbox talk.

The site boundary

Almost everything else in this module depends on the site boundary being set, so do it first on every new job.

Open Site boundary from the dashboard. Pick the project, hit Use site address to drop the centre point from the job’s address, then set:

• Radius (metres) — default 150 m, anywhere from 20 m to 5,000 m. Big enough to cover the site and the parking, small enough that the pub across the road isn’t inside it.
• Strict enforcement — on by default. Sign-in and sign-out from outside the boundary are blocked unless the person gives a reason, which is then recorded against the sign-in.

Watch out: without a boundary, most of this module can’t work for that job. A job with no site boundary can’t be signed into from the app, can’t produce a QR poster, and can’t auto-fill diary weather. If a job seems to be doing nothing in Site Management, this is almost always why.

Tip: the centre comes from the job’s address, so a typo there puts your geofence in the wrong suburb. Worth a look before the crew starts.

Site sign-in

Your team signs in and out from the mobile app — see The Mobile App for what that looks like in their hand. CortexPM records their location, the distance from the boundary, whether they were inside it, and any override reason.

Sign-in records

Sign-in records is the full attendance register. Each row shows who, which site, in and out, and how they signed in:

The method column tells you where it came from: Mobile app, QR gate, Kiosk, or Entered by staff.

People forget to sign out. Bulk close signs out the rows you tick, or — with nothing ticked — everything left open from an earlier day. Closed rows are marked so nobody later mistakes them for measured hours.

The QR gate — signing in people who don’t have the app

Not everyone who walks on needs an account. Deliveries, inspectors, a subcontractor’s labourer for the day.

1. From Sign-in records, pick the project and download its sign-in poster — a branded A4 PDF with the job’s QR code.
2. Put it on the gate.
3. Anyone scans it with their phone camera. No app, no account. They enter their mobile number; if CortexPM knows it they’re confirmed, otherwise they give a name and who they’re with.
4. Any outstanding toolbox talks come up on the same page and are acknowledged there.
5. They’re signed in, and they appear in the same register as everyone else.

The code can be changed at any time. If a poster goes missing or a site changes hands, rotate the code and every printed copy stops working at once — then print a fresh one.

Watch out: the poster is only available once the job has a site boundary set.

Your crew

Crew is where you say who works on which job. It matters for two reasons: it’s how a field worker gets access to a job at all, and it’s what makes the attendance and timesheet numbers line up.

Adding someone. Pick Existing person or Someone new… and give a Name, Trade and Mobile. Anyone you’ve given the Field Worker role already has a profile — the picker lists those first, so you don’t create a duplicate of someone who already has a login.

The mobile number matters. It’s how the QR gate recognises a returning worker, so a labourer who signed in at the gate last month is matched rather than created again.

Linking a gate worker to a login. Someone who’s been signing in at the gate shows a No app login badge. When you give them a CortexPM account, link the two rather than starting fresh — their sign-ins and hours come with them.

Assigning to a job. Set a start date (defaults to today) and leave the end date blank while they’re still on. Removing someone ends the assignment today and keeps the history; re-assigning them later continues that history rather than starting a new one.

Default cost code. Set one per worker and it’s pre-selected when they allocate time. Worth doing for anyone who does the same thing most days.

Watch out: a field worker with no crew assignment sees nothing. Their app says “You’re not on a crew yet” and there’s no job to sign in to. This is the first thing to check when a new starter says the app isn’t working.

Timesheets

Your crew allocate time on the app by tapping what they’re on — they never fill in a timesheet. Timesheets is where you approve what came back.

The screen opens on a week, filtered to Awaiting approval. Other filters: Approved, Sent back, Not submitted, All.

Approval is per worker per day, not per entry — a worker might tap through fifteen segments in a day and you’re not approving those one at a time.

Each card shows three numbers: hours they allocated, hours their sign-in covers, and the difference. Where those disagree by more than half an hour, the day is flagged with an “N h out” badge.

• Approve — done.
• Send back — asks what needs fixing. The worker gets the day back, editable, with your reason attached, and can resubmit.
• Approve all unflagged this week — clears the straightforward days in one go. It skips anything flagged, so the ones worth a look stay on your screen.

Tip: a flag is a prompt, not a problem. Someone who signed in at 6:30 and started allocating at 7:00 will flag every day. Either they tighten up or you accept the half hour — the number just stops it happening silently.

Approving a day also fills the personnel section of that day’s site diary, so the diary headcount and the timesheet agree without anyone typing it twice.

The site diary

One diary per job per day. Quick actions → New site diary, or from Diaries.

Conditions. Fetch Weather pulls the day’s conditions (sourced from the Bureau of Meteorology) using the job’s site boundary as the location, and fills in the condition, temperature high and low, and rainfall. It runs automatically on a new diary. For a past date it pulls the historical record, so a diary written up three days late still gets the real weather.

Headcount. Refresh from Sign-ins counts the unique people who signed in that day. You can override it, but taking it from the register is the point — that’s the number that stands up later.

The record. Work summary, delays, safety notes, visitors, and a Deliveries table (supplier, description, docket number, time, who received it).

Status runs Draft → Submitted → Approved.

Watch out — approval is done on the app. A diary is formally approved from the mobile app by an Owner, Project Manager or Contract Administrator, which stamps who approved it and when, and notifies whoever wrote it. Setting the status dropdown to “Approved” on the web changes the label but does not record the approver — so for anything you might need to rely on, approve it on the app.

Watch out: diary photos and the wet-weather delay prompt are mobile-only. The web form has the free-text delays field, and displays photos uploaded from the app, but has no upload of its own.

Tip: the app’s Insert template button drops a standard structure into the works and safety sections. Get your site managers using it and your diaries start looking consistent across jobs without you writing a policy about it.

Defects

Defects run from the moment someone spots something to the moment you’ve verified it’s fixed — and, if it comes to it, into a formal notice under the contract.

Every defect gets a reference like DFR-00001.

Recording one

Quick actions → Log a defect, or from the app on the spot. You record:

• Location on site and the issue observed
• Severity — Minor, Major or Safety. This sets the order defects are listed in on the register.
• Due date for rectification
• Trade — the cost code it sits under, so you can see defects by trade
• Client visible — whether it shows in the client’s portal and app

Photos and PDFs attach to the defect, and again as rectification evidence when it’s fixed.

Assigning it

Assign to one of:

• an internal user — one of your staff, who gets a push notification, or
• a subcontractor vendor — which routes it into their subcontractor portal and emails them

The assignee list is in two groups — your team and the project’s subcontractors. A subcontractor is always assigned as the company, which is what routes the defect to their portal and their notifications.

Assigning moves a defect to Assigned. Changing the assignee clears any acknowledgement and rectification, because the defect is now with a different party. Unassigning drops it back to Open.

You can bulk assign, bulk set a due date and bulk toggle client visibility from the register.

Getting it closed

A defect moves Open → Assigned → In Progress → Rectified → Closed, and the two ends of that matter:

• Rectified is the person who fixed it saying “I’ve done it” — from the app or their portal, or with Mark rectified on the web. It does not close the defect. It puts it on your verify list.
• Closed is you verifying it. Close out asks what you checked, and that note stays on the record.

If it’s not right: Send back returns it to the assignee with a note about what still needs doing. Reopen does the same for something already closed. Both keep the assignee and prefix the note with who sent it back.

Tip: the register’s to verify counter is the one to watch. It’s work that’s claimed as done but nobody’s checked — the pile that quietly becomes a handover problem.

The register

Defects lists everything, sorted Safety → Major → Minor, then by due date. Filter by status, severity, project, whether it’s overdue, and whether a formal notice has been issued. A filtered view can be bookmarked, so save the views you come back to.

Issuing a formal notice

When a defect needs to go on the contractual record, Issue Formal Notice creates a defect notice as a contract action, with its own PDF, linked to the defect.

CortexPM picks the contract from who’s responsible: a defect assigned to a subcontractor goes under their subcontract; anything else goes under the project’s head contract. If neither exists it tells you rather than guessing — an unassigned defect on a project with no head contract can’t be issued until one of those is sorted.

Closing the defect closes the notice with it, so the informal site record and the formal notice don’t drift apart. See Contracts for what happens to the notice after that.

Toolbox talks

Write the talk — title, content, talk date — and set it active. Active talks are presented to workers during site sign-in, on the app and at the QR gate, and each person taps to acknowledge before they’re on site. The acknowledgement is recorded against that sign-in.

Nobody is assigned a talk individually. Anyone signing in to that job gets any active talk they haven’t already acknowledged, oldest first.

The talk’s detail page lists who has acknowledged it, when, and which sign-in it came from.

Watch out: that list is who has signed off, not who hasn’t. The dashboard’s sign-off percentage is measured against the people who’ve signed in to that job recently, so treat it as an indication rather than a roll-call. For a definitive answer, cross-check the acknowledgement list against your crew.

Tip: make a talk inactive once it’s done its round. Leave five active and every worker gets five screens at sign-in every morning, which is how people learn to tap through without reading.

Safety observations and inductions

Both are captured on the mobile app and reviewed through reporting.

Safety observations — anyone on site can raise a Positive, Negative or Near Miss observation with a description, location and severity. Field workers get the same thing as Near miss / Hazard / Good practice on their Safety tab.

Site inductions are presented and completed in the app, with completion recorded per person.

Watch out: both are mobile-only for now — there’s no web screen for raising or closing a safety observation, or for writing an induction. On the web you see them through the Safety Summary report, which covers observations by severity, toolbox talks and induction completion.

Reports

From the Reports hub:

• Attendance — hours on site per person, company or day, from the sign-in log. Pick a project (or leave blank for all), a date range, and group by Person, Company or Day. Where a worker has approved time entries, the cost-code split shows alongside — so you can see both that someone was there eight hours and what they were on.
• Defects Register — the register as a document.
• Site Diary Summary — diaries over a period.
• Safety Summary — observations by severity, talks and induction completion.

Watch out: Attendance is a record of time on site. It isn’t payroll and it doesn’t interpret an award. Plenty of builders run sign-in for the site record without allocating time at all, and the report handles that — the cost-code column is simply blank.

Who can do what

By default, Owners, Project Managers, Contract Administrators, Site Managers and Field Workers have Site Management access. Accounts, Estimators and Sales don’t. Your company can change that in settings — see Getting Started.

Read-only users see the registers and detail pages but no action buttons.

There’s one extra rule on top of the permission: approving a site diary needs the Owner, Project Manager or Contract Administrator role as well as edit access, so a site manager can’t sign off their own diary. Everything else in the module — including closing out defects and approving timesheets — runs on the plain edit permission.

Field workers are a special case. They get a purpose-built app rather than these web screens, limited to the jobs they’re crewed to, and they never see a dollar figure — no rates, no budgets, no totals.

See also: The Mobile App for what your site team and crews see · Contracts for defect notices · Subcontractor Portal for how subcontractors receive and respond to defects · Jobs & Job Cost for the job address that drives the site boundary.