Honest Take — Module 8: Runway Math, Burn, and Threshold Decisions #

The honest framing for this module is that you have been making your biggest career and money decisions emotionally — "can I leave this job?", "can I drop this client?", "can I hire?", "can I invest more in the product?" — and the runway model converts them from gut-feel to mechanical. I do not mean that the model removes judgment; it doesn't. What it does is make the judgment apply to inputs you actually understand, rather than to a vague holistic gestalt that mixes financial pressure, career anxiety, ambition, and time-since-last-deposit-cleared into a single uneasy feeling. The model separates the inputs. It says: given your liquid assets, your monthly inflow from each stream, your monthly outflow including taxes and household and entity costs, your runway under the worst case is X months and under the likely case is Y. With X and Y on the table, "should I drop the anchor client?" becomes "what runway do I retain after dropping them, and is it above my written threshold?" That is a different question, and the different question has a defensible answer.

If a live decision is in front of you right now — an offer in flight, a contract renegotiation, a quit-and-build itch — run the model against each realistic outcome this week, before the decision point. The offer converts at target; the offer comes in low and you negotiate; the offer doesn't come and the current configuration continues; the anchor client ends and the next one takes two quarters. Each scenario has a different optimal action, and the model is what makes the optimal action visible in advance. The act of running the scenarios ahead of the moment is itself a posture intervention: you walk into the conversation knowing you can decline a weak outcome, which is the M5 BATNA-as-runway connection closed — your negotiation walk-away strength is literally a row in this spreadsheet. This is also where I should name how the two earlier drafts of this curriculum differed: one carried this module whole, as its distinctive contribution; the other treated runway inside its emergency-fund and entity-buffer material. The merged edition carries the full module because the distinction matters — the emergency fund (M4) is defense; the runway model is agency.

The income side of the model is more complex than it looks because each stream has a different cash-flow shape. A retainer is monthly and contracted. Product revenue is small but predictable, or lumpy and uncertain, depending on the product. Consulting is milestone-based and clusters. A salary, if one arrives, runs on payroll cycles. A stream that produces the same annual total can land evenly or all in one quarter, and runway is a per-month calculation, so the model needs monthly granularity with the actual shape of each stream. If you pay quarterly advance tax (the Indian independent operator does; many jurisdictions have equivalents), the tax outflows are large and lumpy, and a model that smooths them will overestimate the trough months. Explicit tax rows, actual timing. The expense side is where most engineers underestimate honestly, because the household base number is uncomfortable to write down precisely — the recurring floor (rent, utilities, school fees if applicable, parents' contribution if applicable, insurance premiums prorated), the variable layer that absorbs more than anyone remembers, and the entity's operating drip hidden in credit-card noise. Write it all down. The discomfort is the data; the engineers who look away from the actual monthly number are the ones who later wonder why the runway feels shorter than the gut estimate suggested.

A note on behavioral scaffolding, because the bare spreadsheet doesn't hold everyone's behavior. The cash-buckets discipline — separating money by purpose into distinct accounts so that the tax money, the buffer money, and the spendable money cannot visually blur — is the load-bearing idea inside Profit First, and you can take the idea without the book's US-banking mechanics. If you've noticed that a single large balance makes you feel richer than you are, the buckets are for you; if the spreadsheet alone keeps you honest, skip the ceremony. The principle is the same one as the two-pool model in M7: separation at the account level is what makes the numbers legible at the decision level.

The threshold conditions are what the model produces and what makes future decisions easier. Written in cold blood: "I drop the anchor client when side revenue exceeds X for Y consecutive months and runway is at least N months." "I make the first hire when product revenue exceeds W and the hire's cost is below V% of net margin, with buffer to be wrong for six months." "I raise capital when..." — and for a one-person product company the honest answer is almost always never: equity is the most expensive money there is, bootstrapped optionality is the entire point of the runway you're building, and the threshold for revisiting is a specific winner-take-most opportunity that exceeds your runway capacity. The fourth threshold — "I invest in the product at a higher pace when..." — is the one most people forget to write, and it matters because under-investment is also a decision: side work that could compound sits at maintenance pace not because the math says so but because no one ever wrote down what number would justify acceleration. Once the thresholds are written, the future decision becomes mechanical: the metric crosses the line, the action fires — robust against both varieties of self-deception, quitting on a good month's euphoria and never quitting out of un-modeled fear. Without thresholds, every career decision is an incident handled live at 2 a.m. With them, the decision flow was decided in daylight.

Define the worst case honestly, because the model is only as useful as its ugliest scenario. For the contractor archetype, the honest worst case is not "revenue dips 10%"; it is "the anchor client ends and the replacement takes two quarters" — a scenario that has happened to enough independent engineers that treating it as a tail event is self-flattery. If the model can't survive being run against that scenario without you flinching, that is information about the model's job, not a reason to soften the scenario.

Conclusion #

Build the twelve-month cash-flow model with monthly granularity, explicit tax-outflow rows, and the actual shape of each income stream. Run it against the live scenarios in front of you this week, not after they resolve. Write the threshold conditions for your versions of drop-the-anchor, first-hire, and raise-or-not — the third of which is almost certainly "no, at this scale." With the model in place, the big decisions become mechanical against documented thresholds, and the model becomes a standing agenda item of the M13 monthly review. The model is the operational deliverable; the practice of updating it is the durable skill.

Predictions #

The first time you write down the actual monthly household spending number, it will be 20-30% higher than your gut estimate. That gap is not a moral failure; it's the visibility deficit the model exists to close.
The runway under the honest worst case — anchor client ends, nothing converts, current side revenue only — will be tighter than you expect. That tightness is exactly why you should not be anchoring low in whatever negotiation comes next.
The expense audit will surface at least one subscription you forgot you were paying. Cancel it; the saving is small, the audit hygiene is the larger benefit.
You will be tempted, at least once, to optimize the model rather than maintain it — scenario layers, refined assumptions, dashboards. Resist. The model is decision support, not an engineering aesthetic project.
The drop-the-anchor threshold will land at a revenue level you are not currently at but can see from where you stand. Knowing the number will change the urgency and direction of your side work within a month of computing it.
If a live offer or renegotiation lands after you've built the model, you will notice yourself less needy in the conversation. That posture shift is the model paying for itself before any threshold ever fires.
By month four of monthly updates, you will catch a drift in real time that the gut-feel regime would have missed. That single catch will justify the entire effort.