Reflection — An Honest Take 8 min

Honest Take — Before You Begin

Honest Take — Module 5: The Honest Roadmap — Shape Up, Cycles, and the Graveyard #

Shape Up will land easier than Torres or Klement did, and I want to predict why before you confirm it: Ryan Singer writes for engineers in a way most PM authors don't. Bounded scopes, explicit appetites, hill charts, a vocabulary that maps cleanly onto how you already think about systems. You will read the appetite-not-estimate chapter and feel a small click — the click of a framework speaking your language. If you have spent years estimating tasks and watching the estimates miss, the appetite frame is the one that makes the missing not your fault: you don't ask "how long will this take," you ask "how much time is this worth," and you ship whatever fits inside the answer. Hold onto the click. This is the most operationally accessible module in the curriculum for a working engineer, and also the one whose discipline gets tested fastest.

Here is the prediction I want pinned. Your first cycle will not run cleanly. Whatever you ship in the first 4-6 week cycle will leak past the appetite, probably by 30-50%, and the two earlier drafts of this module disagreed about what the leak is. One read it as the standard engineer-estimation error you have made for years and will keep making — partly scope creep, partly under-shaping — and said the second cycle improves not because you learn to estimate (you won't) but because you learn your actual appetite-vs-output ratio and start shaping smaller. The other draft read the week-4 moment as the founder's trap: you are tempted to extend the cycle because killing or shrinking the work feels worse than slipping the deadline, and slipping is, in the long run, more expensive than killing. I think the first draft explains cycle one and the second explains every cycle after it.

Either way the rule is the same: ship at the appetite boundary even if "incomplete," or kill the cycle. Do not extend. The skill is in the cycles, not in the document — read Singer once, run cycles five times.

The unflattering pattern worth staring at: scope creep has probably killed more of your portfolio than lack of demand has. The asymmetry is visible in most engineer-founders' public output, and the maintainer archetype shows it cleanest — the open-source package with dozens of releases ships small slices reliably for years, while the products sit at various pre-revenue states. Same person, same skill, different scoping regime. When the work is bounded (a release, a changelog entry), things ship; when the scope is open ("the MVP," "the launch"), things don't. Shape Up is the rate-limiter that converts the second mode into the first, which is why this module sits before metrics, pricing, and launch: nothing downstream matters if nothing ships.

The graveyard is the most important artifact in this module, and I want to be specific about why. A roadmap is what you intend to do, and intentions are cheap. A graveyard is what you decided not to do, and decisions are expensive. Every entry represents a feature or direction you considered, evaluated, and rejected with reasoning, and the reasoning is the asset: when a future "we should build X" arrives — a customer asks, a competitor ships it — you return to the graveyard, read why you rejected X before, and decide whether conditions actually changed or whether you're being seduced by recency. Without the graveyard, every request feels new and urgent. Aim for twelve entries minimum. If you can't find twelve things you've decided not to do, you have not been deciding; you've been deferring, and the discomfort of pushing from entry eight to entry twelve is precisely the work.

One cultural note and one cross-reference. The book's register — opinionated, slightly preachy, certain of itself — will rub you the wrong way in places. Fine. The methodology is portable; the swagger is not the methodology; use the technique and ignore the attitude. And when, mid-cycle, a new product idea arrives dressed as inspiration, apply M4's test before touching it: would you put a current survivor on the kill list to make room? The cycle boundary is also a commitment device against your own starting reflex.

Conclusion #

Shape Up is the most engineer-accessible PM methodology in print and the module where shipping behavior actually changes. Your first cycle leaks; your third runs close to clean; the skill is the cycles. Scope creep — not lack of demand — is the failure mode that has been killing your portfolio, and the bounded appetite is the rate-limiter that fixes it. The graveyard converts vague restraint into recorded decisions. Read Singer once; run cycles five times; ship at the boundary.

Predictions #

  • The appetite-not-estimate frame will click within the first 30 minutes of reading. Note where you were when it clicked; the memory is useful later when the discipline gets hard.
  • Your first shaped pitch will be under-shaped — too much "we'll figure it out," not enough "this rabbit hole is explicitly out of scope." The second will be sharper.
  • The first cycle will leak past appetite by 30-50%. Ship at the boundary anyway and learn from the gap; do not extend the cycle.
  • At week 4 of the first cycle you will seriously consider extending. Track the consideration in writing — it is evidence about you, independent of what you decide.
  • The graveyard will start with 4-5 easy entries and stall. Entries 8 through 12 will take real effort to surface and will be the most valuable ones.
  • You'll be tempted to bring story points or velocity tracking back into the cycle "for visibility." Refuse; Shape Up rejects them for reasons that apply doubly at solo scale.
  • The hill chart will feel silly until you've used it for two cycles, then feel essential. You cannot skip to "essential" by reading about it; run it badly twice.