Building your plan step by step
Sixty-eight chapters supplied the materials. This one is the build. It lays out the full sequence that turns them into a plan with numbers in it, tested, written down and ready to run. A plan that does not live on one page does not exist (Choosing your strategy: criteria, comparison, worked case: without that page you have an intention, not a strategy).
There are seven steps, and their order is not up for debate. The numbers before the target, the target before the portfolio, the portfolio before the rule, and all of it before you write anything down. Each step points to the chapter that covers it in full; this one is the parts list and the thread. Budget two to four evenings for a first pass, one of them entirely for step 6, the writing, the step everyone skips and the one that actually makes the plan.
The final product is in a callout below: the template for the one-page plan, the document you, your spouse and your 2043 self should be able to read in five minutes. The chapter closes on day one. A validated plan gets run, through the transfers, the orders and the automation that move it off the page and into your life.
Step 1: the numbers of your life (one evening, the most important one)§
It starts with three numbers you audit rather than estimate (How much you need). Spending first: twenty-four months of statements, lumpy items annualized, health coverage added at full cost. Split it into a floor you have tested (the five-year test, better still the trial quarter, Flexibility: myth and reality (what it can really absorb)) and the comfort level you are aiming at. Then your pension: Social Security or an employer plan, the statement rerun with no further earnings, marked down 10 to 20% for the risk that the promise moves, with the amount and the start date written down. Then the horizon: the 90th percentile for the last survivor, plus two to four years for the socioeconomic gradient (Horizon, life expectancy, and 50-year retirements). Last comes the inventory: what you own, listed account by account, the near-certain side income marked down (Real estate in a FIRE plan (your home, and rentals), Pensions and other income in the plan), and an estimate of your blended effective tax rate.
Step 2: the target and the date (one hour)§
Pick the multiple (How much you need): 25x to 33x, depending on the horizon, on what the market costs today (Valuations, the CAPE, and what they say about your withdrawal rate) and on the margins you carry. The target follows, and so does the date, projected at your current savings rate. Write the departure criteria right away, while calm, with the anti-goalposts clause and the date-or-condition commitment (The one-more-year syndrome). They hold best when you set them years ahead, while nothing is at stake.
Step 3: the portfolio (one evening)§
The method behind it, designing around risks rather than around assets, is in Designing a portfolio: the method, not the model. Here we follow the order the book used. The equity share first, set by the two tests: a drawdown you can live through, and a floor that stays covered. The plateau forgives a lot of imprecision (The stock and bond allocation in retirement). Then the contents: cap-weighted world equity, with any tilts written down (International diversification (and home bias), Factors (Fama-French, value, momentum) in the withdrawal phase), and a defensive sleeve drawn from the table of regimes (intermediate nominals, linkers, a dose of gold, long bonds or trend depending on what your audit says, Defensive assets: the map and the roles, Market regimes (growth × inflation, sticky bears) and why they matter). Then the time dimension, the bond tent: bring the equity share down with new money starting now, and write the climb back up into the plan (Glide paths: the bond tent, rising equity, and the fragile window). Last, where it all sits: the shopping list of blocks, and which account holds each one (Building it with US-listed ETFs: the retiree's shopping list, US accounts, and the order you drain them). Six to eight holdings, each with a named role.
Step 4: the rule and the protections (one evening)§
Set the withdrawal rule with the five-step procedure (Choosing your strategy: criteria, comparison, worked case: the admission tests, the matrix, the hybrid by phase), and put a number on every parameter. Most plans land on the same shape: steered through the bridge years, proportional once the pension covers the floor. Next the buffer and its flow rules, meaning size, trigger, refill and the year it goes away (The cash buffer: size, cost, and what it is really for, Drawing on a buffer and refilling it: the rules that work). Then the ladder, if you are building one for the bridge floor (Bond ladders (and the ladder of linkers)). Then flexibility, in writing: an indexation freeze, a cut at a threshold (Flexibility: myth and reality (what it can really absorb)). Last, the margins you name without counting them: skills kept current (Barista, Coast, side income: work you choose), a home you could turn into cash (Real estate in a FIRE plan (your home, and rentals)), whatever side income comes naturally.
Step 5: validating across models (one session with a simulator)§
The finished plan goes on the test bench, with the audited inputs and nothing else. A typical session is walked through control by control in Using the FIRE simulator; here is the recipe that closes the build. Read it on the weight of the evidence: the central case acceptable at your threshold (Failure probability: reading it, choosing it, and not letting it run you), the broad sample tolerable with its failures examined (Historical windows, bootstrap, parametric: the three families of models), the 1966 and 2000 vintages crossed, a lost decade survivable. Then look at the standard of living actually delivered in the worst quartile of paths, in the depth of the cuts and in how long they last, not just at the failure probability. The gap between the central model and the sequence stress tells you what a bad sequence would cost, softened by your protections (Sequence of returns risk: the retiree's real enemy). Sensitivities settle the rest: spending plus or minus 10%, returns plus or minus 0.5 points, the pension two years late. Does the decision survive all of it? A solver, working in equivalent moves, is the last cross-check: it converts each margin into dollars, into years or into flexibility (if a cheap lever is lying around, take it now). A plan that passes this recipe is a plan. The others fail one test and go back to the step they came from. That is the build working as intended.
Step 6: writing it down (the evening everyone skips)§
Filling in the template is not paperwork. It is the final design test. Every box that resists points to a decision you never really made: the threshold you cannot put a number on, the rule you cannot summarize. The finished document then becomes the infrastructure for everything else. It carries the anti-panic (The psychology of spending: why spending it is the hard part: the rule written while calm), the governance of a couple (FIRE as a couple, and as a family: the reverse quiz is run on this page), the continuity of the plan if you are not there (your letter of instructions points to it), and the guardrail against yourself (the never-dos of block 6 are the most reread lines in a storm). Date it, version it, and file it where your spouse will find it.
Step 7: day one (one morning)§
Doing it is mechanical. The automatic transfers first, the plan's monthly paycheck (The psychology of spending: why spending it is the hard part). Then the orders at your broker: a gentle migration to the target portfolio, using new money and the accounts where trading costs you no tax first, and spreading the taxable sales over several years. The buffer funded (The cash buffer: size, cost, and what it is really for). The vehicle for side work opened, even if it stays dormant (Barista, Coast, side income: work you choose). The paperwork: beneficiary designations reread, the letter of instructions, the power of attorney. And the first review booked in the calendar. The plan is alive.
The essentials§
- Seven steps, one order: audited numbers, then the target and the departure criteria (written now, while calm), then the portfolio (the split, the blocks, the accounts, the tent), then the rule and the protections (every parameter numbered, flexibility in writing, margins named), then validation on the weight of the evidence, then writing it down (the one-page template, the real deliverable), then day one (the automation, the paperwork, the calendar).
- One decision per step, written down with its one-line reason, and no going back before the full pass is done. An imperfect v1 beats an eternal v0.4.
- The eight blocks (numbers, portfolio, rule, protections, thresholds, never-dos, governance, margins) fit on one page. A block that overflows means a design too complicated for its weakest executor.
- Writing it down is the final test. The box that resists is the decision you never made. The document is the infrastructure of the anti-panic, of the couple, and of continuity.
- The plan is alive. It is born complete, it runs from day one, and after that it changes in the one place built for it: the annual review (The annual review: the retiree's checklist).
Going further§
- Every step points to its own chapters: this book is the documentation for the template.
- The validation session, control by control: Using the FIRE simulator. Picking the rule: Choosing your strategy: criteria, comparison, worked case. The life of the plan afterwards: The annual review: the retiree's checklist and When to worry, and when to let it ride.
- The worked households running through the book (Aline and Thomas, Claire and Idris, Léa) are the fastest way to see the seven steps hang together.