Before you begin
- Math ladder: R11, R12, R13, R14, R15, R16, R17, R18, R19, R20, R21, R22

Lesson 01 · Section 1 of 8 · Reading
Every one of the 36 modules in this domain runs a fixed formula. This chapter is those formulas — all twelve operations they perform between them, taught from zero.
Every one of the 36 modules in this domain runs a fixed formula. This chapter is those formulas — all twelve operations they perform between them, taught from zero.
It is the highest-leverage chapter in the course. Not because it's the hardest, but because it is the same twelve ideas over and over. Once you have them, a module stops being a puzzle and becomes a recognisable shape: oh, this is a chain with a wall in it or this is an interior optimum. Twelve ideas unlock thirty-six modules.
Worth repeating, because this is the chapter where people expect to be caught out.
The engine does every calculation. What changes here is that you'll understand why the engine's answer is what it is — which is what lets you predict it, argue with it, and know when a number should surprise you.
Chapter 3 gave you everything this chapter assumes: you can read a rate, spot a weighted average, keep units straight, and reject an impossible number. Nothing more is needed.
Twelve is too many to hold. So they come packaged as five headline ideas, each with one anchor figure worth carrying:
| # | The idea | Anchor | Rungs |
|---|---|---|---|
| 1 | Chains multiply — and the tightest link sets the number | 70 deals from 5,000 leads | R11, R12 |
| 2 | Money has a clock — later is worth less, and churn compounds | $360k over three years is worth $298.4k | R14, R16 |
| 3 | Uncertainty has arithmetic — a chance times a value is a number | 30% × $50k = $15k | R17 |
| 4 | Only the next step matters — decide at the margin; things saturate | the best split returns $130k | R18, R19, R22 |
| 5 | The best answer is usually in the middle | the peak is at 2.0×, worth $60k | R20, R21 |
Two more rungs sit inside those ideas rather than getting their own slogan: R13 (margin — what a discount really costs) and R15 (decomposition — splitting a total into signed parts).
| # | Piece | Rung | Modules it unlocks |
|---|---|---|---|
| s01 | The revenue equation | R11 | all diagnosis |
| s02 | Multiplying a chain | R11 | sdx-03, pql-04 |
| s03 | Why improving a chain is multiplicative | R11 | sdx-03 |
| s04 | Headroom versus leak | R7, R11 | sdx-03 |
| s05 | A chain with a wall | R12 | sdx-01 |
| s06 | Why the loud stage isn't the binding one | R12 | sdx-01 |
| s07 | Splitting a total into parts | R15 | sdx-02 |
| s08 | Signed contributions | R15 | sdx-02, ngx-04 |
| s09 | Margin, and what a discount really costs | R13 | dst-01, wtd-01 |
| s10 | Fixed, variable, and break-even | R20 | dst-03/04 |
| s11 | Compounding survival | R16 | rnx-01, dst-01/03 |
| s12 | Why money later is worth less | R14 | dst-02 |
| s13 | NPV in plain arithmetic | R14 | dst-01/02, enb-01 |
| s14 | Probability, and what 30% means | R17 | pql-02, rnx-02 |
| s15 | Expected value: the workhorse | R17 | pql-01/02/03, sdx-04 |
| s16 | Total versus marginal | R18 | enb-02/03, scd-03 |
| s17 | Diminishing returns and saturation | R19 | 9 modules |
| s18 | Finding the optimum | R20, R21 | ngx-01, scd-01/02, enb-01/04 |
Then D05 — Math Drills (four short stations, about three minutes), and a recall set.
Allocation under a budget (R22) is taught at the end of s16–s18 as those three ideas applied at once, because that is exactly what it is.
Most of this chapter can be skimmed or taken out of sequence. The last three cannot. Marginal → diminishing → optimum → asymmetry → allocation is a single chain where each step is computed from the previous one's output:
Skipping ahead doesn't make it harder. It makes it impossible, because the number you'd need doesn't exist yet.
Every worked figure in this chapter comes from a real module's engine, re-derived from that engine's own formula. Twelve modules are quoted: sdx-01, sdx-02, sdx-03, dst-01, dst-02, rnx-01, pql-02, scd-01, scd-02, enb-02, enb-03, enb-04.
That's deliberate. You are not learning simplified teaching examples that later turn out not to match — you're learning the actual arithmetic of the actual modules, one chapter early.
Fourteen traps, and they come in two kinds. Knowing which kind you're facing is half of not falling for it.
Outcome traps (eleven). The naive choice loses money, and you can see how much. Fix the worst-converting stage returns $350k where the right fix returns $1,400k. Fix the loud stage returns exactly $0.
Reading traps (three). The naive number is real, usually bigger, and simply isn't the quantity the decision turns on. Throughput is 190 where the average capacity is 318. A rep's total coaching value is 100 where their next block is worth 0.
Those three are the harder kind, because there's no loss to point at — only a number answering a question nobody asked.