Section 1
The matrix
The single number that drives the whole matrix is margin per constrained hour: the profit a job produces divided by the hours it consumes of your scarcest resource, usually your certified or expert people. Not margin per job. Not revenue. Margin per constrained hour. A job can be highly profitable in absolute terms and still belong in the decline-first quadrant if it ties up your ceiling-setting people for days to earn what a better job earns in an afternoon.
Section 2
Why "profitable" is the wrong filter on its own
Most owners triage by one question: is this job profitable? Under a capacity ceiling that question is close to useless, because nearly everything is profitable in isolation. The job that fills a week and clears a decent margin looks fine on its own. It only looks wrong when you notice what it displaced: three shorter, higher-margin-per-hour jobs you had to turn away because that one long job ate the week. This is the trap the matrix exists to break. When capacity is fixed, accepting a job is not a standalone yes. It is a choice against every other job those same hours could have done. The right filter is therefore comparative rather than absolute. The question is no longer whether this job is good. It is whether this job is the best use of the hours it will consume. A firm that says yes to everything profitable ends up jammed with mediocre-per-hour work, turning away its best jobs by accident, because the calendar was already full of lesser ones.
Section 3
Scoring a job in ten minutes
You do not need a spreadsheet with twenty inputs. You need four numbers per job and one division. 1. Estimated profit on the job. Price minus your true direct cost. Be honest about the cost, including the parts-running and callbacks a messy job invites. 2. Constrained hours consumed. Not total hours. The hours of your genuinely scarce people, the ones who set your ceiling. A job that keeps a labourer busy for a week but your master electrician for two hours consumes two constrained hours, not forty. 3. Margin per constrained hour. Divide 1 by 2. This is the job's score. 4. Strategic weight, up or down. Adjust for what the raw number misses: a job for an anchor client, a job that opens a lucrative segment, or conversely a job with reputational or callback risk. Keep this adjustment small. It is a nudge, not a licence to override the number every time your gut prefers a job. Rank your live and incoming jobs by score three, place them in the quadrant, and act. The discipline is in following the matrix when a decline-first job is from a customer you like, or a qualify job is bigger and shinier than a protect job. The number is right more often than the gut, especially when the gut is tired and the calendar is full.
Section 4
A worked scoring, so the quadrants are not abstract
Put three real jobs through it. A domestic consumer-unit swap: profit 400, and it uses 4 of your master's hours, so 100 per constrained hour. A commercial fit-out: profit 6,000, but it ties up your master for 60 hours, so 100 per constrained hour as well. And a fault-finding call-out on a contract you already hold: profit 350 for 2 of your master's hours, so 175 per constrained hour. The commercial fit-out is by far the biggest number on the invoice, and most owners would grab it on sight. But per constrained hour it scores the same as the small domestic swap and well below the short call-out. If you take the fit-out, you have committed sixty of your master's scarce hours to a job that earns the same per hour as work you could do in an afternoon, and for those sixty hours you cannot take any of the higher-scoring call-outs. The fit-out is not wrong to take. It is a Qualify job: take it selectively, price the scarcity hard, and never let it crowd out a month of Protect work. The point of the scoring is that the invoice size told you the opposite of what the constrained-hour math tells you, and under a ceiling the constrained-hour math is the one that governs.
Section 5
The two models under the matrix
Comparative statics: shift the mix, not the volume. Under a fixed ceiling you cannot change how many hours you have. The only variable you can move is what those hours are spent on. The matrix is a comparative-statics tool for your job mix: hold capacity constant, change the composition toward higher margin per constrained hour, and total profit rises with no new headcount. This is the lever that works when the hiring lever does not. Comparative statics, the mix lens. Assumes you can shift composition while holding capacity fixed. Fits because job mix is the one input a capacity-ceilinged firm actually controls. Breaks when your work is undifferentiated and every job scores the same, which is rare but possible in pure commodity trades. Counteracts the reflex to grow by volume when volume is the thing you cannot grow. May reinforce over-pruning, so keep enough Fill work to cover the gaps between your best jobs. Opportunity cost: every yes is a hidden no. The matrix makes the shadow price visible. When you accept a decline-first job, the cost is not just the hours. It is the Protect job you could not take because those hours were gone. Owners consistently underweight this because the job they turned away never appears on the books. It is invisible, so it feels free. The matrix forces the invisible cost onto the page by ranking every job against the alternative use of its hours. Opportunity cost, the allocation lens. Assumes a real alternative use exists for the constrained hours. Fits because a full calendar guarantees that yes to one job is no to another. Breaks in a genuine lull when no better work is competing, at which point Fill work is correctly accepted. Counteracts the habit of judging jobs in isolation. May reinforce arrogance about future demand, so decline based on the pipeline you actually have, not the one you hope shows up.
Section 6
Making it a Nordic and DACH-aware discipline
In markets where subsidies or stimulus are inflating demand, the matrix matters more, not less. A renovation-grant boom or a stimulus-fed order surge fills your inbound with jobs faster than you can serve them, which is precisely the condition where undisciplined triage does the most damage. When demand is artificially high, the temptation to grab every subsidized job is strongest, and the cost of jamming your ceiling with decline-first work is highest, because the Protect jobs you displace are also more plentiful and more lucrative in a boom. The busier the market makes you feel, the more the matrix earns its keep. Run it hardest when the phone is ringing most.
Section 7
The blind spot
The matrix optimizes the allocation of hours you have. It is blind to two things. First, it assumes your estimate of constrained hours per job is roughly right, and a job that surprises you by ballooning can turn a Protect job into a hidden decline-first job after you have already committed. Build a margin of error into the constrained-hours estimate for unfamiliar work. Second, and more fundamentally, the matrix takes the ceiling as fixed and only rearranges what sits under it. It does nothing to lift the ceiling. Triage keeps you profitable inside your constraint. It never removes the constraint. The only levers that add capacity, systematizing the work and growing your own people, live in the clone-the-founder and grow-your-own playbooks in this cluster. Triage buys you the margin and the breathing room to fund those slower bets.
Section 8
The fitness test
Take the three jobs currently eating the most of your scarce people's time. Score each on margin per constrained hour and compare them to the last three jobs you turned away. If any of the jobs you declined would have scored higher than the jobs you are now grinding through, you have been triaging by accident, taking whatever called first and letting it block your best work. Score every job for one month before you accept it. If your revenue rises with no new hire, the ceiling was never the only problem. The allocation was.