{{first_name}} Between 2004 and 2007, I worked as a Sr. Accountant at Carillion Constructions’ Canadian operation, in the road maintenance division.
Carillion was a giant. A global construction company headquartered in the UK that grew to more than £5 billion in revenue; hospitals, highways, railways, government contracts across three continents, tens of thousands of employees. I sat inside their finance function for three years. I saw how a company at that scale bids, books, and reports its work.
On January 15, 2018, Carillion collapsed. Not restructured. Not sold. Liquidated. The largest construction failure in British history. On the day it died, a five-billion-pound company had roughly £29 million of cash left. Thousands of subcontractors were left holding unpaid invoices.
Here's what should keep you up at night: Carillion wasn't killed by fraud, or a recession, or one villain in a boardroom. It was killed by something every contractor reading this does every week.
Bad Bids.
Here is the story, in plain English. Carillion won enormous fixed-price contracts. Two hospitals, a major highway by bidding thin to keep revenue growing. Then it booked the profits it expected to make on those jobs years before the money was real. When the jobs went wrong; design problems, delays, costs running past estimates, reality arrived all at once. In July 2017, the company admitted a £845 million write-down on problem contracts. That single announcement vaporized years of reported profit. Six months later, the company didn't exist.
A handful of badly estimated fixed-price jobs. That's what it took to kill a five-billion-pound company. Now let me show you why the same math is pointed at yours.
On a fixed-price job, your upside is capped and your downside is not. If you execute the job perfectly and you make your margin; eight, ten, twelve percent. That's the ceiling. But if the estimate turns out to be wrong; labour runs 30% over, site conditions surprise you, a supplier reprices mid-job then there is no floor (except bankruptcy in certain cases). A job can lose 40 or 50 percent of its contract value. No job ever makes 50 points. Carillion's contracts had ceilings and no floors. Perhaps, so do yours.
Let’s take a small contracting business for an example. A $5M-a-year company running 8% net makes $400K in a good year. One $800K fixed-price job that goes badly wrong and loses $350K erases nearly a full year of profit. Every good decision, every clean job, every well-run crew, wiped out by one bad estimate.
And earning it back is worse than losing it. At 8% net margin, recovering a $350K loss requires roughly $4.4 million of new work; bid correctly, staffed right, executed flawlessly. A year of perfection just to get back to zero just because of one bad bid. Ten good jobs cannot repair what one bad one destroys. That arithmetic doesn't care whether you're a £5 billion company or a $5 million one. I've now watched it work at both ends of the scale.
Where do the bad bids come from? Carillion's were the giant version of the same four I see in businesses your size:
Estimating from hope instead of data. If you don't know what your last twenty jobs actually cost versus estimate, every bid is a guess wearing a hard hat. Carillion booked hoped-for profits; small contractors bid hoped-for productivity. Same disease.
The desperation bid. Carillion bid thin because it needed revenue growth to satisfy the market. You bid thin in a slow season to keep crews busy. Work that loses money doesn't keep a company alive. It spends the company's savings to stay busy.
Unpriced scope. Changes absorbed to avoid friction. Small each time. Together, often the entire margin.
The unfamiliar job. The worst losses almost always come from a company's first attempt at something. It could be a new building type, new contract size; priced with the confidence of their tenth.
What Can You Do About This:
The defenses are unglamorous.
A written bid/no-bid rule you apply before falling in love with the revenue. Contingency sized to this job's risk, not habit. No change order starts without a signed price. And a twenty-minute autopsy on every completed job, estimated versus actual so the next estimate learns something. Carillion had none of that discipline at £5 billion. You can have all of it at $5 million.
The Number You Can Use:
$4.4 million. That is the new revenue required, at 8% net margin, to recover one $350K job loss. Run the formula on your own business: any job loss, divided by your net margin, equals the flawless new work needed just to get back to zero. Most owners have never done this division. Carillion's board apparently hadn't either.
One question for you this week: What did your worst job actually cost you and did your estimating system learn anything from it, or did only you?
Why am I writing this? I've spent over 20 years working alongside construction and trade business owners across Canada. My firm, N3 Business Advisors Inc., has helped hundreds of contractors buy, grow, value, and sell their businesses. I started my career as a teacher, and that part of me never left. So once a week, I put one lesson in your inbox, for free, to help you build a business that buyers line up for. If this was useful, pass it to one owner who needs it. If it wasn't, reply and tell me what you'd rather read about.
Nitin Khanna, CFA · Founder, N3 Business Advisors · Mastering the Business of Construction