The Order Book Is Full. The Margin Came In Thin.
Monday starts before the shift does. Two jobs from last week are back on the floor.
The estimator has eleven quotes out and nobody can say why the last twenty-four died.
Somebody wants an answer on the hot job, and the schedule that answers it is still in your head.
You are not short of work. You are short of the margin the work was priced to carry, and the part of your plant that could produce more is already paid for and standing idle.
Fair warning: if cutting is always your first move, this page will annoy you. The most expensive leaks in a shop never show up on the cost line.
See the Full System →up to 1.2%
Canadian manufacturing growth expected in 2026, below historical potential, with Ontario down 50,000 manufacturing jobs since January 2025.
~95%
Enterprise GenAI pilots that produced no measurable impact on profit. Not that the tools failed. Nothing reached the profit line.
69%
Employees in the United States and Canada who are not engaged (52%) or actively disengaged (17%). The structural variable most growth investments do not account for. The global equivalent is 80%.
Five Symptoms Converge on One Revenue Leak.
Five questions. Five symptoms. One leak between the margin you bid and the margin you keep.
The Pattern Applies in Manufacturing the Same Way It Applies Across Sectors.
The manufacturing twenty, installed on top of the universal eighty. Stabilise the bleeding, systematise the operating layer, scale to the proven outcome, documented over a 100-day operating sequence.
Five Questions. Five Minutes. If Even One Makes You Pause, Keep Reading.
-
01
The margin you bid is your strategy. The margin you realize is your system. When they disagree, which one are you scaling?
-
02
Your top salesperson, your senior foreman, your best PM. What does each one get rewarded for hitting? And do any of those rewards push them away from what the business actually needs right now?
-
03
Your top three customers account for a share of revenue you can probably guess to within five percent. Now the second number: how much of that share was won by you personally, and how much by the business as a system?
-
04
Name the manufacturing operating model problem you've tried to fix three times in the last five years. The crew, the consultant, the software, or all three. Why is it still here?
-
05
If you had to commit to a new strategy tomorrow, a different customer to aim at, a different offer, a different sales motion, which would be the hardest to move? The people, the processes, or the technology you've installed? Which answer surprised you?
If you can answer all five in 30 seconds, the rest of this page isn't for you. If even one made you pause, keep reading.
Two Manufacturers, Every Word the Same, Two Different Businesses.
Two manufacturers can use every one of the same words and win revenue in completely different ways. Before anything else on this page lands, it is worth knowing which one you are.
Line, changeover, spec, yield, order book, shop floor. Both of them say all of it. They do not make money the same way, and a fix built for one does not land in the other.
This page follows the left column, because that is where I have installed and where the numbers on it come from.
If you are in the right column, the pattern still holds and the install is a different job. Tell me which one you are and I will tell you what I would look at first. I have not published a case in the right column and I am not going to imply otherwise.
Owner-Led, $5M to $50M. In Manufacturing It Lands Hardest Higher Up.
We work with owner-led businesses between $5 million and $50 million, in manufacturing, construction and technical professional services. That is the band and it belongs in the open.
In manufacturing the work lands hardest between $25 million and $45 million, and the reason is structural rather than commercial.
At that size there is an organisation to install into. Five or more people carrying a real number, an estimator, a scheduler, a plant manager. Separate seats mean the handoffs between them are real handoffs, and the handoffs are the whole job. The back end is also large enough that cost comes out while the front end is speeding up. Two levers running at once inside one engagement.
Below that, the owner is usually standing in the seam himself. The pattern still holds. There are simply fewer separate seats to install between, so the install is a different shape and a shorter one.
This is a preference, not a gate, and it is a statement about where the work is most effective rather than where the buyers are. It is also scoped to the left column above, because that is where it was observed.
Five Symptoms. One System Gap.
The five questions aren't five separate problems. They are one revenue leak showing up in five places. Half of it is arithmetic: what you bid against what you realize, what you win against what you keep, what AI cost against what AI returned. The other half is people: what your team is rewarded for, what your seniors are stretched across, and what your foremen do when the schedule slips. Treating either half alone is why most growth advice fails. Treating them together is the system we install.
The trade-off is real, and it's worth naming. The cost-leadership path squeezes every dollar out of every job: lean operations, tighter bids, faster cycles. It works until the macro turns against you, which it did in 2025. The premium-revenue path compounds margin instead of squeezing it: a named customer you serve on purpose, a system that runs without the founder, and an engaged team that turns AI into leverage instead of noise. Both paths are valid. bEffective installs the second.
If your strategy is to grow, your operating system has to do something your old one didn't. That's what we build.
Six Things the System Has to Be Told Before It Is a Manufacturing System.
Roughly eighty percent of what a manufacturer needs is the same thing every business needs, and it is written up on the System page. The pattern of rooms and seams is on Operations. This is the other twenty.
One
The quote is the entire Sales room, and it is an engineering document.
The price is set by an estimator working from a drawing, often before anyone commercial has spoken to the customer. The seat that owns the price is not the seat that owns the customer, and in most shops those two seats have no standing meeting.
Installs: a margin rule at the estimating desk, a recorded reason for every quote that loses, and a feedback loop from actual job cost back into the estimate.
Can your estimator tell you, without opening a file, which of last quarter's quotes lost on price and which lost on lead time?
Two
Backlog is the forecast. Pipeline is a different instrument.
A manufacturer manages an order book with promised dates. A funnel with probabilities measures something else. Installing pipeline stages on a shop that runs a backlog produces two systems and one truth, and the owner ends up trusting neither.
Installs: the order book as the primary revenue instrument, with the quote log feeding it rather than competing with it.
When you say you know what next quarter looks like, are you reading the order book or a forecast someone built beside it?
Three
The delivery leak already has a name, a ledger and a department. Nobody has joined it to revenue.
Manufacturing calls it cost of poor quality. It is the only one of the three sector words that already has a measurement discipline, a trained person and a filing cabinet behind it. And it sits in quality, not in revenue. The scrap report and the revenue report are written by different people for different meetings, and the arithmetic joining them is never run.
Installs: the quality ledger read as a revenue ledger. Scrap, rework, quality escapes and changeover loss expressed as the margin they removed from jobs already won and already priced.
What did last year's scrap cost you in margin, not in kilos?
Four
Capacity is the currency, and recovered capacity is revenue you already paid for.
On a plateau below 78%, the growth argument is not sell more. It is stop buying capacity you already own.
That plateau figure is Statistics Canada's, from the final release of a programme they have since discontinued, and roughly a third of it is imputed rather than reported. I use it because it is the best available, and because a plateau is a diagnosis rather than a verdict. You should know how soft it is before you build anything on it.
Installs: capacity recovered from the hidden factory counted as revenue capacity, before any conversation about a machine.
The last time you costed a new machine, did anyone cost the capacity you are already losing?
Five
The RFQ-to-floor chain is four handoffs with no seam owner.
RFQ arrives. Estimator prices it. Job is won. Scheduler slots it. Floor builds it. Shipper ships it. The promise made at the quote has to survive all four intact. Each department looks fine on its own. The money disappears in between. That is the Operations pattern wearing manufacturing clothes, which is why this page does not rebuild it.
Who owns the space between the quote and the schedule?
Six
Customer concentration is your Demand room.
A manufacturer's demand is not a market. It is a handful of OEM customers, a tender cycle and a position in somebody's supply chain. The Demand question is not how many leads. It is what happens to the year if the largest customer re-sources.
If your largest customer re-sourced in January, which month would you feel it?
A Leak Is Not a Percentage of Revenue. It Is a Percentage of Your Profit.
A leak worth X% of revenue costs you
X divided by your net margin, of your profit.
A 1% leak on a 10% margin business takes 10% of profit. The same 1% leak on a 4% margin business takes 25% of profit. Same leak, very different year.
It is division, so it cannot be argued with, and every reader arrives at his own number rather than ours. That is why there is no recovery percentage anywhere on this page. Your margin decides what a leak is worth, and I do not know your margin.
What the delivery leak runs at, in your vocabulary.
Cost of poor quality averages roughly 15% of the sales dollar in manufacturing, with a range of 5 to 35% depending on product complexity.
Source and objection. Institute of Industrial and Systems Engineers, summarising J.M. Juran, Quality Planning and Analysis, third edition. It is a textbook estimate summarising practitioner experience rather than a sampled population, and it is undated. Treat it as a baseline, not a measurement.
And what happened when somebody actually measured both halves.
A 2011 study tracked quality costs inside a single manufacturer using both the conventional model and a hidden-cost method. Visible quality costs came to 5.64% of sales revenue. Hidden quality costs came to 8.78%, which is 1.6 times what the conventional model captured. Combined, 14.42% of sales revenue.
The paper's own conclusion is the line worth carrying: the company's total quality costs exceeded its current profit margin.
Source and three objections. Cheah, S.J., Shahbudin, A.S.M. and Taib, F.M. (2011), "Tracking hidden quality costs in a manufacturing company: an action research," International Journal of Quality and Reliability Management, Vol. 28, No. 4, pp. 405 to 425, published 19 April 2011. One. It is one case company, so it is depth rather than breadth. Two. It is action research, meaning the researchers took part in the implementation. That is a recognised method and it is not an independent audit. Three. A secondary source describes the case as a wooden manufacturer while the paper's own abstract says continuous-process, so it is described here as the paper describes itself and the industry is not characterised.
You do not have a capacity problem. You have a hidden factory.
Armand Feigenbaum, who ran quality at General Electric, named it: the part of your organisation that exists to do bad work. He put it at 15 to 40 percent of capacity.
Every point of that you recover is a point you do not have to buy.
Where the Margin Went
Busy quarter, thin margin. The three places money leaks between the price you agree and the cash you keep, and three comparisons you can run this week.
Different Vocabulary. Same Leak.
Between 2003 and 2005 I worked as a subcontract operator at ProGuide, a North American management consultancy running enterprise-grade revenue operations. I was not deploying my own system there. I was observing and using theirs, at the $1B client scale where every workstream had structural rigour. One of those engagements mapped revenue and operational architecture at Canadian Blood Services and found $10M of cost inside a single process workflow, by reconciling what sales had committed to against what operations could actually deliver.
The engagement that proved the discipline scales down was DEW Manufacturing, an aerospace fabricator at roughly $50M. Same reconciliation, applied across the manufacturing workflow, and it doubled profits across the engagement. I was a subcontract operator on that work, not the accountable operator, and that distinction matters: the system I install today is the one I first watched run at enterprise scale and then learned to compress to SMB without losing the discipline.
Different vocabulary. Same leak.
The five questions above don't change when the industry changes. Neither does the answer.
Two Manufacturers. One I Was Accountable For, One I Was Not.
Tier 1 · Accountable operator of record
Zohar Group
A Canadian polyethylene packaging manufacturer, quote-driven, with sales scattered across every order that came in and margins compressing. The commercial team was measuring roughly 150 accounts per representative. Sales was rebuilt around the segments that carried margin.
30% revenue growth in the first year
40% lift in lead conversion in the first year
15% cut in production costs in the first year
Tier 2 · Contributing operator
VT Controls
An industrial controls and building-automation integrator. The sales strategy was built for recurring services; operations billing was still set up for one-time capital sales. Reconciling the two was one piece of a wider engagement.
$4M to $9M over 24 months
25:1 return on the engagement fee
I guided the work. I did not install it, and that distinction is the point of this case.
Why the Zohar number moved, because a figure without its mechanism is just a number.
An account count is an activity count. Roughly eighty percent of the revenue came from twenty percent of the accounts, which means the large majority of what was being counted produced nothing measurable. The team was busy against a denominator that did not matter.
It is the same error as counting machine hours and calling it output. You already know what that looks like on a floor. It looks the same in a sales meeting.
Questions Manufacturing Owners Actually Ask.
How does a $25M to $45M manufacturer fix owner-dependent revenue?
By moving the decisions out of your head and into seats that already exist. At that size you have an estimator, a scheduler and a plant manager, so the work is not hiring. It is writing down the rules each of them is currently guessing at: the margin rule at the estimating desk, the promise-date rule at the schedule, and the escalation rule for the hot job. The test is not whether you could leave for a month. It is whether the answer to the hot job comes from the board rather than from you.
What does a broken quote-to-build handoff actually cost?
It costs the difference between the margin you bid and the margin you kept, and it costs a different amount in every shop. A leak worth 1% of revenue takes 10% of profit in a 10% margin business and 25% of profit in a 4% margin business. Same leak, very different year. That is why the arithmetic on this page runs on your margin rather than quoting a recovery percentage.
What if my price comes off a price list rather than a quote?
Then revenue is won in a territory and through a channel, and the handoff diagnostic on this page will not grip. The leak is more likely to sit between list price and net realised price, after discounts and rebates, rather than between the quote and the job cost. The pattern holds. The install is different. There is no published case in that column and this page will not imply one.
We are smaller than $25 million. Does this still apply?
Yes, and the install is different. Below that size the owner is usually standing in the seam himself, so there are fewer separate seats to install between and the work is shorter and more concentrated. The band on this page is where the work lands hardest, not a gate.
Is this the same as hiring another salesperson?
No, and on a capacity plateau it is often the more expensive answer. Another salesperson adds quotes to a desk that already cannot say why the last twenty-four died, and adds jobs to a floor that is losing margin on the ones it has. The question is what happens to the margin on the work you already won, before you go and win more of it.
What is the difference between cost of poor quality and a revenue leak?
Arithmetic, and which meeting it gets discussed in. Cost of poor quality is already measured in most shops, by a trained person, in a ledger, inside the quality department. A revenue leak is the same money expressed as the margin it removed from jobs you had already won and already priced. Scrap in kilos goes to the quality meeting. Scrap in margin goes to yours.
How long does an install take, and what happens when you leave?
The Blueprint diagnostic takes two to four weeks. The Build takes 100 days from kickoff, and most outcomes start showing up in the third or fourth month. What happens when I leave is the point of the engagement rather than a footnote to it: I install the system, run it until it runs without me, and hand it over.
Go Deeper
Canonical · The Beginning
What Is a Revenue Blueprint?
A priced diagnostic you own outright: ten areas mapped in three groups, every leak priced per year, portable to any operator.
Read the argument →
Tool · Five Minutes
Revenue Leak Calculator
Slow follow-up, quiet churn, soft pricing, and rework nobody logs. Five inputs, one conservative estimate of your annual leak.
Run your numbers →
Bring one finished job and one quote that lost.
Pick a job you completed in the last year and find the margin you bid and the margin you kept. Then pull one quote you lost in the same period and find out, from whoever would know, whether it lost on price or on lead time. Three answers, all of them already inside your own business. Most owners have never had them in front of them at the same time.
- What you leave with: where the fade opened on that job, whether it opened at the quote, on the floor or at the billing, and the one number that would tell you on the next job whether it closed. In writing.
- How long it takes: thirty minutes, and it ends at thirty minutes.
- What it is not: no deck, no proposal, no ERP recommendation, and nobody asks to see your books. If you want any of those afterwards you can ask for them.
- What we do first: we read the published capacity and margin figures for your sub-sector before the call, so the time goes on your job rather than on our background.
- When not to book it: if you have not pulled the numbers, there is nothing to read and the thirty minutes becomes a conversation about manufacturing margins in general. Pull them first, or leave this alone.
The operator on the other end of that call is Mike Reardon, four decades running and fixing revenue systems, and the accountable operator at United Systems Solutions, Zohar Group, Diabsolut and Dale Carnegie across four eras and four industries. The engagements are written up here, numbers, dates and all, including the ones where the role was contributing rather than accountable.
This isn't for every manufacturer. The ones who want the margin they bid to survive the floor, yes. The ones whose answer to a thin year is to cut another line item, there are better fits.