Skip to content

Working copy

Draft text, rebuilt from the manuscript on every change. Stubs and source passages are still here, and none of it is final. Back to the working copy contents

Drafted The chapter is written and still being revised.

The trades: who holds the jig

The clock on the wall of the bid depository is not an ordinary clock. It prints the time on whatever is pushed into it, and on bid day the estimators of a dozen mechanical and electrical firms take turns pushing envelopes into it as late as they dare.

The scene is invented, but the clock is not. On jobs that use a bid depository, a trade contractor does not send its price to the general contractors bidding the job. It sends it to the depository, sealed, stamped and on time, and the general contractors collect the prices from there. In Yukon, the depository's clock is the official time, and a copy of every bid is kept for at least a year. The estimator who finishes early does not hand in early. A price that leaves the office early can find its way to a rival, or be shopped by a general contractor to a rival, and the rival only has to beat it by a little. Trades dislike being shopped, so they hold their prices until the last useful moment.

So everyone hurries to be last. It looks like bad planning from the outside. It is a protocol the trades helped build, out of an envelope and a clock, to keep a price where it was put.

A tradesperson would call that a jig.

The jig

A jig is a device that holds the work and the tool in the right relation to each other, so that a hard or dangerous cut can be made again, safely, by anyone on the crew. The best jigs are often not bought. The person who knows the cut builds one, and the setting stays theirs.

This chapter carries that one image, because the trades are the part of construction that already knows how to make one. The depository is a jig for prices. The machines that may one day arrive on site could be jigs too, or they could be something else, and the difference matters more to the trades than to anyone else in this book.

"The trades" here means three groups. Trade contractors are the firms that price and build one part of the job, mechanical, electrical, concrete, drywall, and bid to the general contractor rather than to the owner. Tradespeople are the people who do the work with their hands, many of them in unions. And the trades have their own institutions: the contractor associations, the depositories, the unions and the training bodies. Each meets this project at a different distance.

One level down

Chapter 4's map has the owner tendering to a general contractor, and the general contractor competing on price. That fight is real, but it is not where most of a building's price is made. The general contractor builds its bid from the trades' prices, so most of what a building tender contains was priced by people who never see the owner. The hard bidding still happens, one level down.

That has a consequence for any machine that reads a tender. When a bot misreads the drawings, the error does not stay where it was made. The general contractor's reading and the trade's reading can differ, and the gap between two trades' work, the duct nobody priced because each thought it was in the other's scope, is the oldest argument on a job site. The cost of a misreading runs downhill. Trade contractors are also near the back of the line when a contractor fails. They are usually the last ones left holding it.

One limit is worth saying plainly. The project's price method will read public bids, the prime bids that owners publish. Trade prices are private, and the project has no data on them. Whatever this chapter says about the trades' own prices comes from people, not from the method.

More people manage than build

Two sets of figures, from two countries, point the same way.

In Australia, between 2006 and 2025, the number of construction managers went from about 61,000 to about 131,000. Over the same years the number of trades workers went from about 345,000 to about 414,000. Managers more than doubled; trades grew by about a fifth. In 2006 there were about five and a half trades workers for every manager. By 2025 there were about three.

In the United States, between 2005 and 2024, the trades went from 71 percent of the construction workforce to under 59 percent, while managers went from 10 to 17 percent.

More of the industry now manages and documents than builds. Why is not known. New design and project software, stricter codes and compliance rules, bigger firms, and perhaps job titles that changed name all play a part, and nobody has measured how much each does.

There is another way to read the same numbers. By most measures, output per construction worker has been flat for decades, or worse, and the trades' share of those workers has fallen. If output per worker has only held flat, each tradesperson is now carrying more of the office on their back. The front line has been getting more done per head, somewhat more than the records show: on the US figures, something like a fifth over nineteen years. That is not a dramatic number, and nobody can yet say whether it came from better tools, parts made in factories or the managers' own planning. But it turns the usual story around. The trades are not the part of the industry that lags. They are the part that quietly got better.

It also tells the project where to aim. A second opinion on a price is office work. So is reading a tender, checking a quantity and answering the last week's questions. If machines can take some of that weight, the place it comes off is the office layer that has been growing on the trades' shoulders, not the front line. The bidders' line in this book is that they would like to spend more time building and less time on the paperwork of a proposal. It is the trades' line before it is anyone else's.

One caution belongs right next to that hope. Time saved in an office tends to be filled by new tasks. A check that adds forms is not a gift to anyone, and this chapter does not promise the trades fewer managers.

What a cleaner tender gives a trade

What a trade gets from the project is mostly indirect, and better said that way than oversold.

A cleaner tender reaches them too. Fewer contradicting addenda, fewer questions left unanswered in the last week, and scopes split between trades in a way that is written down rather than assumed. Each question a designer answers before the tender closes is one less argument on site months later, and the trades are the ones who have that argument.

A second opinion before the envelope closes helps a larger trade contractor bidding public work directly, where its price meets the owner's estimate head on. A small firm whose work is simple and familiar probably needs it less. That is fine. Not every tool has to fit every hand.

What the trades will fear

The fears are better named here than said back to the project later.

The first is faster bid shopping. A general contractor's machine that compares every sub-bid in seconds is the depository's old enemy with a new engine. The depository's answer still holds: seal the price, stamp the time, keep a copy. Anything the project builds should work with the clock on the wall, not around it.

The second is being watched. A machine that reads everything can feel like a machine that watches everyone. This is why the third far world of this project is called bot admin, not bot monitoring: what the machines keep track of is the work and the documents, not the people.

The third is replacement. Machines are coming to construction sites, slowly, and the trades have heard the promises before. That is where the jig comes back.

The jig, grown up

Most of today's site robots are set up away from the people who work beside them, by programmers at head office or at the maker. A jig is the opposite. The person doing the work makes it, sets it and owns it.

This book's hope is that the machines arrive as jigs. It even has a word for one:

A jigbot is what a tradesperson builds when they set up and teach a robot for the task at hand, the way they'd build a jig.

The idea has a history. Just after the Second World War, some machine tools could be taught by a machinist's own hands: the tool recorded the movements and played them back. They lost out to machines programmed in an office, away from the shop floor. The jig is that road not taken, offered again.

Setting up a jig makes you its author. It should not make you its insurer. If the teaching was sound and the machine still went wrong, the bill should not land on the person who set it up simply because they were the last one to touch it. The trades already live with this arrangement for their tools: training, a ticket to show for it, and rules on who answers for what. A machine that learns needs the same three things, and the trades' own training bodies are the people best placed to write the ticket.

Two honest limits. A jig only repeats, and the machines people picture would do more than repeat; nobody yet knows how a crew keeps the setting in its hands when the machine learns. And a jig holds still, while the work between a person and a machine will be a back-and-forth. The image is not perfect. It is the best one the trades have offered so far, and it is theirs.

Not everyone will hold the jig. Some workers will set the machines up. Some will end up assisting them, which is the arrangement unions will rightly worry about most, if it comes with less skill and less pay. Some may find in a patient machine the teacher they never had. The last chapter of this book comes back to where each of them might stand.

Where the trades could start

The trades do not need to wait for robots to have a say. Three things could start now, one for each group.

A bid depository could ask whether its clock and its kept copy would serve as the outside witness for a machine's reading of a trade bid, and what its rules would need to say if they did.

A trade contractor's estimator could take a past sub-bid on a public job and mark where the tender's split between trades was unclear. That gives the project its first look at the scope gap from below, without asking anyone for a price.

And the unions and training bodies could start the conversation about who sets the machines up before the machines arrive, while the ticket is still a blank page.

The trades have built their own protocol once already, out of an envelope and a clock. They are the people best placed to say whether a second one fits the same hand.