Where office admin time
actually goes.

The fix for an office chasing site paperwork all week is to capture each record once, on site, as it happens, on one record everybody can reach. Nobody can tell you how many hours your office spends on admin, and any published figure is invented. Measure it for one week with a tally of four categories: assembling positions, retyping what site already recorded, chasing paperwork that exists but cannot be found, and correcting errors caused by the first three.

Measure yours rather than borrowing a figure

There is no shortage of statistics claiming that construction firms lose some large percentage of their week to administration. None of them is traceable to a source worth naming, and none of them describes your business.

Which is fortunate, because measuring your own is genuinely easy and takes a week. A tally sheet with four columns, filled in by the two or three people who do most of the office work, marking a tick in fifteen minute blocks. Not a time and motion study. A rough count that is nevertheless about you rather than about an average.

Firms that do this are usually surprised in a specific way: the total is not the shock. The distribution is.

The four that dominate

Consistently, and in this rough order.

1. Assembling a position that already exists in pieces. The monthly cost report, the application, the answer to a director asking where a job stands. The information exists across the accounts package, a spreadsheet, an inbox and somebody’s memory, and the work is not producing it but reconciling it. This is nearly always the largest single category.
2. Retyping what site already recorded. Hours from a paper sheet or a photograph of one. Delivery notes. Dayworks. Plant movements. The record was made once on site and is being made again in the office, with a delay and a transcription error rate attached.
3. Chasing paperwork that exists. Not missing documents; findable ones. An insurance certificate somebody has, a signed variation in an inbox, a delivery note in a van. The document is not lost, it is merely not anywhere anybody else can reach.
4. Correcting the consequences. The application queried because the measure was unsupported, the payment held because a certificate expired, the invoice disputed because nobody could find the instruction. Each of these is generated by one of the first three.

The largest category is almost never doing the work. It is reconciling four versions of information that was correct when it was first recorded.

Putting a number on it

Once you have the week, the arithmetic is short. Take the hours in each category, multiply by the all-in cost of the people doing them, and multiply by fifty. The all-in cost is not the salary, for the reasons in what an employee actually costs you.

Then do the more uncomfortable calculation: what those people would be doing instead. In most contracting businesses the office is not overstaffed, so hours recovered do not become redundancies. They become tendering that was not happening, retention that was not chased, applications submitted on time, and a director looking at a current position rather than a three week old one. That is where the value actually is, and it is larger than the wage figure.

What actually reduces it

Two things, and neither is working harder.

Capture once, at the point it happens. Almost every hour in category two exists because a record was made on site and then made again in the office. Where the site record is the record, that hour disappears rather than being made faster.

Hold the position in one place. Category one exists because the answer is distributed. It shrinks in proportion to how much of the picture lives on one record, and it is the reason integration questions matter more than feature questions when firms compare systems, which is covered in choosing trade and construction software.

Category three is a permissions and location problem: things people can reach without asking. Category four largely disappears when the first three do, because it was never independent.

What not to conclude

Two things, because this measurement is easy to misuse.

It is not a judgment on the people doing the work. Nobody in a contracting office is slow. They are assembling a position from four systems because that is the only way it can be assembled, and they are usually doing it well.

And it is not automatically a case for buying software. If the total is small, the honest answer is that you do not have this problem yet and the money is better spent elsewhere. The measurement is worth doing precisely because it can tell you no.

What contractors move to

Three routes, and the tally tells you which one the problem needs.

1. Better discipline with the tools already in place. One shared folder per job, a weekly cut-off for site paperwork, and sheets photographed the day they are signed. It cuts chasing and leaves retyping where it is.
2. Off-the-shelf job software. Timesheets, diaries and forms captured on a phone and reported from one system. It removes most retyping where the firm's paperwork matches the product's forms.
3. A bespoke platform. The firm's own forms, approval routes and reports on one record, so the office assembles nothing that site has already recorded.

Where to start, on Monday

Print four columns, hand them to whoever does most of the office work, and ask for a tick in fifteen minute blocks for one week. Assembling, retyping, chasing, correcting.

At the end of the week, look at the distribution rather than the total. If assembling and retyping together are most of it, you have a structural problem that more effort will not fix. If they are not, you have a different problem and this was a cheap way to find out.