Somewhere in your office right now, an hours total that a shifter wrote down at the face has been typed into at least three separate places. The daily cost sheet. The weekly hours tally. The payroll run. And when the invoice goes out, someone pulls those same numbers one more time to build the backup the client requires.
The number has not changed. It has just been copied four times from four pieces of paper, by one person, with no automated check between them.
That is the construction timekeeping problem for every mining contractor running a paper chain: not capturing the hours, which the shifter handles fine, but moving them from that capture to a payroll run and a client invoice without introducing errors at every step. A mismatch at payroll is a wage dispute. A mismatch at the invoice is a client dispute. A mismatch between the two is your problem to untangle after both have already been paid.
Worked example
A 12-person underground crew working 7 days on / 7 days off, two shifts per day. The pay period runs 14 days. The office manager handles the hours tally, payroll entry, and invoice backup from field timesheets.
| Step | Calculation | Result |
|---|---|---|
| Shift entries captured per pay period | 12 people x 14 days = 168 shift entries | 168 shift entries |
| Re-entry events downstream of initial capture | Each entry moves from field timesheet to (a) daily cost sheet, (b) weekly hours tally, (c) payroll system, and (d) invoice backup: 168 x 4 = 672 re-entry events | 672 re-entry events |
| Office time at an estimated 3 minutes per re-entry event | 672 x 3 = 2,016 minutes; 2,016 / 60 = 33.6 hours | 33.6 hours per pay period |
| Office cost at $45/hr fully loaded | 33.6 x $45 = $1,512 per pay period; $1,512 x 26 pay periods = $39,312/year | $1,512 per pay period / $39,312 per year |
Answer: On a 12-person crew, rekeying the same shift hours across four records absorbs roughly 34 hours per pay period. At $45/hr fully loaded, that is about $39,300 a year in office time before a single correction, query, or disputed shift.
Where the number starts
A shifter at the end of a production round writes down three things: who was on the crew, what they worked on, and how long each task took. Eight hours drilling, four overtime, one and a half standby waiting on ventilation. The writing takes five minutes and the information is correct.
That is the last easy moment.
The form comes off site with the crew. It may have passed through two pairs of hands, been photographed and sent via WhatsApp, or arrived in an envelope with the vehicle pool. The office manager deciphers it - a different skill from accounting - and enters it into the daily cost sheet. Entry two.
At the end of the week, the same hours move to the weekly hours tally, where she checks the additions against the posted schedule. Entry three. At the end of the pay period, those hours go into payroll. Entry four. If the client wants backup attached to the invoice, she pulls the same numbers one more time. Entry five.
No single step is hard. All five say the same thing. And yet the number that left the face correctly can arrive at the invoice incorrectly, because one person read the same handwriting four times under four different circumstances and once got a digit wrong.
Three records that have to match
The timesheet is not just a payroll input. It is the record that has to reconcile with three other things at once.
The daily cost report. Hours on the timesheet should match hours on the daily. They often do not, because the daily tracks task time against cost codes and the timesheet tracks total person-hours against a name list. A shifter who wrote "VA" on the daily used a code the office enters as "VENT" in the cost sheet. Same shift, same hours, different label. Reconciliation by hand, shift by shift, every week.
The payroll run. Payroll applies a different logic to the same inputs: straight time, overtime, rotation premiums, stat holidays. A timesheet that shows only a total forces the payroll processor to reconstruct which hours attracted which rate. On a 12-hour shift crossing a stat holiday in a seven-on rotation, that reconstruction is not straightforward, and an error here is a wage dispute.
The client invoice. The invoice goes out with the billable hours. But not all hours on the timesheet are billable. Standby time you paid because the mine's ventilation was delayed may or may not be claimable under the contract. If the timesheet does not carry that split from the field, someone has to make the billing call at invoice time, from a reconstructed account of why the crew was waiting two weeks earlier.
Three records. One source. Rebuilt three times. When a client disputes a line on your invoice, it is usually not because the work did not happen. It is because the paper trail is spread across a daily report, a payroll file, and a spreadsheet, and the three of them do not quite agree.
What construction timekeeping software changes about the chain
The fix is not a new input at the face. The fix is one input that feeds all three outputs without being retyped.
Construction timekeeping software captures the hours once, at entry, and routes them to the daily cost sheet, the payroll processor, and the invoice backup without anyone touching them in between. The office manager does not transcribe. She approves.
That distinction matters for two reasons. First, approval is the work that requires judgment: deciding whether a standby claim is billable, flagging a shift that does not match the posted schedule, questioning a week with more overtime than the crew normally runs. These are decisions, and decisions cannot be automated. Transcription is a pipeline that a person is running because the pipeline has no software in it yet.
Second, when the software handles the chain, the office manager starts at the exceptions instead of spending 34 hours per pay period moving numbers between files before she can see what the exceptions are. She does not lose her role. She loses the part of the job that was always a waste of her time - and the company's money.
This matters most when the crew moves between contracts in a single pay period, which happens regularly in mining contracting. One person, two sites, two cost codes, two billing rates. A manual system almost always introduces a gap in that split. A single-entry system carries the cost-code flag from the field entry through to the invoice without anyone reconstructing it.
No signal underground
The question comes up immediately: there is no signal at the face. How does a digital timekeeping chain work underground?
The same way it has always worked for initial capture: on paper. A driller filling out a field book at 600 meters is not going to stop because a software vendor needs a tablet entry. Nothing changes at the face. The brief for every underground operation should read the same way: capture happens on paper, by hand, exactly as it does now.
What changes is what happens to that paper when it reaches surface. A photo sent via the site's surface Wi-Fi when the crew portals out, or a scan transferred by USB from the shift supervisor's phone, feeds into a system that reads the form rather than a system that requires a human to retype it. The form is the same. The re-entry disappears.
The lag between when a shift happens and when the office has a readable record of it is where most underground timekeeping errors originate. A question about a missing depth measurement that reaches the driller at 19:00 on shift day gets answered in a text message. The same question surfaced at payroll submission two weeks later gets answered from memory, if at all. Closing that lag is the variable the technology controls, not the interface at the face.
The reporting chain cost calculator at bigdogmining.com lets you run the exposure for your own operation: crew size, pay period length, and the time each re-entry step currently adds. The number it returns is what the paper chain is costing before anything is automated.
The payroll-is-two-days objection
The most common pushback from owners is that payroll only takes two days per pay period, so there is nothing broken.
That is accurate about the payroll step. It is not accurate about the chain.
Those two days run on inputs that have been accumulating and being retyped for 12 days before them. If the hours arrived in the payroll system clean - one entry per person per shift, time-stamped, cost-coded, with the overtime flag already set - the two days stay two days.
If the hours arrived from three spreadsheets with slightly different task codes, including a week-3 correction to a week-1 standby entry that was miscoded at the time, those two days turn into four. Not because payroll is hard, but because the input is inconsistent.
The investment is in the chain, not the payroll run. Payroll is the last step. The waste lives at step two, where the same correct number gets retyped for the first time.
What the records law requires
The Fair Labor Standards Act requires employers to retain payroll records for at least 3 years. Those records include wages paid, hours worked, the basis for the pay rate, and the calculation of overtime for each workweek. The specific requirements are at 29 CFR Part 516 on ecfr.gov.
For a mining contractor running multiple crews on rotation across multiple sites, this requirement has a practical edge. A spreadsheet updated in place, without a version history, is difficult to reconstruct 18 months later when a former employee files a wage complaint. A timekeeping system that logs each entry, each edit, and each approval produces that record as a side effect of normal operation.
The 3-year retention window also means that years of daily records you already have are worth something as a baseline. They show an auditor what your hours looked like consistently over time. That baseline only holds if the records are legible and reconcilable across the whole span, which a stack of photographed field books in a SharePoint folder is not.
Standby: the hours you pay but cannot bill
Standby is where paper timekeeping fails most quietly, because the error is invisible until the invoice goes out.
A crew waits two hours for ventilation. The shifter writes "SBY 2hr" in the field book. That entry survives all four re-entries and reaches payroll correctly. The crew is paid, and that part works.
Whether the standby is billable is a different question, and the answer lives in the field at the moment the delay happened. The contract specifies that standby is claimable when the mine causes the delay, at the posted standby rate, with a signed delay form. The shifter who wrote that entry knew whether the mine caused it. That knowledge does not transfer automatically through four re-entries.
By the time the invoice goes out, the office is working from a two-week-old field note with no billing flag on it. Someone makes a call from memory. The client disputes it. The contractor has no contemporaneous record to point to.
A billing flag on the standby entry at capture is one field: billable or absorbed. The shifter fills it in at the face, where the information exists. The office manager confirms it when she approves the timesheet. The invoice assembles the billable standby total without anyone reconstructing it at month end.
This is the category of hours where the exposure is hardest to measure and easiest to avoid. Standby you paid and absorbed as a contractor cost is a real cost. Standby you paid and failed to bill because the field record never carried the distinction is a choice you made without knowing you made it.
Run your own numbers
Rather than work it by hand every time, the reporting chain cost calculator does this arithmetic and shows its working. The rest of the set is on the mining calculators page.
Common questions
Why am I hand-adding 8 plus 4.5 plus 1 plus 1.5 to make 15, for every person, every day?
Because the timesheet was built to hold totals and the daily was built to hold tasks, and nobody connected them. The office adds the task hours to produce the timesheet total because the two records have never been the same document. That repeated arithmetic is a symptom: the underlying problem is that the same shift lives in two places with no shared format between them.
What do I do when the timesheet hours do not match the daily report hours?
Trace back to the field entry. One record was filled at the face when the information was fresh; the other was filled later or from memory. Find which one was original and use it as the source of truth. Document the correction with a date and reason. The gap itself tells you which point in the chain is introducing the drift.
What is the right pay period for a seven-day two-shift rotation?
Fortnightly aligns cleanest with the seven-on seven-off cycle: the pay period covers exactly two rotation cycles, so no rotation splits across a period boundary. That alignment removes a common source of overtime miscalculation, because the system sees the full seven-day block rather than counting hours that wrap from one period into the next.
How do I handle standby hours that we pay but cannot bill?
Track the billing flag at capture, not at invoice time. The shifter on site knows whether the delay was mine-caused or contractor-caused. That distinction belongs on the field record when it is made, not reconstructed at month end. Standby you pay and absorb is a real cost. Standby you pay and fail to bill because the field record is ambiguous is an avoidable one.
Payroll is only two days a period. It is not worth automating.
Payroll takes two days because those inputs have been accumulating cleanly, or not, for the preceding 12. The automation is not a payroll tool. It removes the re-entries at steps two and three. When inputs arrive clean, payroll stays two days. When they arrive inconsistent, the two days absorb the cleanup that should have happened 12 days earlier.
Getting this off paper
Most contractors already know these numbers. The cost is in re-keying them into a report, then a timesheet, then an invoice. If that is where your week goes, book a 20-minute call and we will look at your actual paperwork.
