Reflection — An Honest Take 8 min

Honest Take — Before You Begin

Honest Take — Module 10: Up the Stack #

A flag before anything else, because it matters more here than anywhere else in the curriculum: I have never been in an executive meeting. I have not had a CEO check their phone while I was presenting; I have not delivered a 1-pager and watched it come back with red pen on it. The advice in this module is filtered through the writings of people who have been in those rooms — Bezos's memo culture, Minto's pyramid lineage, the staff-engineer canon — not through my own experience of the medium. That matters because executive communication is unusually embodied: the phone glance, the closing body language, the moment a listener shifts from absorbing to deciding. Those signals are read in real time, and the reading builds only in the room. The structural advice here is well-tested and translates directly; the embodied calibration needs a human who has sat at the receiving end of bad engineering presentations. When you start spending time in those rooms, find that person.

The structure itself is an inversion, and naming it plainly is most of the module: engineers present evidence-first because that's how proofs work — here's the context, here's the data, here's the chain of reasoning, and therefore, conclusion. Executives process conclusion-first — here's what I recommend, here are the three reasons, drill into whichever you doubt. Present evidence-first to a decision-maker with four other meetings that hour and they are lost by minute two and deciding without you by minute eight. The pyramid is not dumbing down; it's packet ordering for a receiver with a brutal latency budget. The deeper meta-skill is the one most engineers never reach: the signal you've internalized this module is not that your presentations land — many variables decide that — but that you stop being surprised when they don't. You start recognizing, in the moment, that the conclusion wasn't in the first sentence, that the ask was buried, that the room went closed at minute eight. Recognition precedes correction, and most engineers never even reach recognition.

A warning about the translate-to-business-consequence table, because it can be overdone into self-sabotage. There's a failure mode where an engineer learns the trick and starts rendering every technical concern in inflated business terms — "the codebase is fragile" becomes "this is a five-million-dollar risk" on shaky arithmetic. Executives detect this immediately and discount everything you say afterward. The discipline is to find the actual business consequence and state it accurately even when it's smaller than you wished: "this costs us roughly two engineer-weeks per quarter" is credible; "this is bleeding the company" is noise. The technique works exactly when it's honest. One earlier draft of this curriculum carried Klaff's frame-control material here — the observation that pitches fail when the listener's frame (what they think this conversation is about) is misaligned with yours: engineers default to information-frame, executives to risk-frame, customers to outcome-frame. The concept is load-bearing; the book's packaging is exhausting. Take the one idea and leave the bravado.

Two applications beyond the boardroom. First, interviews: senior-role interviews are increasingly exec-shaped — a time-boxed listener asks a question and decides, from your answer, whether to keep going. "What's the hardest technical decision you've made?" should be answered like a CEO asking about a technical investment: conclusion first ("we chose X over Y, and I think it was right"), three reasons, evidence on demand — not "so back in 2021, we were working on this project, and..." Pair this with Module 5's 90-second rule and you've closed most of the senior-signal gap. Second, open source: if you maintain anything with corporate users, you'll eventually face exec-shaped conversations wearing casual clothes — a sponsor leaning on the roadmap, a company wanting support terms. A written one-page response travels through their organization (their CTO, their procurement) in a way your verbal "no, that's not the direction" never will. Treat important maintainer decisions like executive communications.

The three-lengths discipline deserves a defense, because it looks like over-preparation and isn't. Preparing every proposal at 30 seconds, 3 minutes, and 30 minutes is really an admission about executive time: you do not control how much of it you'll get. The meeting gets cut in half; the decision-maker arrives late and leaves early; the elevator actually happens. The engineer with only the 30-minute version delivers the first ten minutes of context and gets cut off before the recommendation. The engineer with the 30-second version delivers the conclusion and the ask in any window the day actually provides, and expands only if invited. The compression also has a side effect that is the real point: if you cannot say it in 30 seconds, you don't yet know what you're asking for.

And never skip the follow-up note. Within the hour: decisions made, owners, dates, "tell me if I've captured anything wrong." Executives operate across so many threads that the meeting's conclusions genuinely decay within a day, and whoever writes the recap quietly becomes the author of record for what was decided. It's five minutes of writing for a disproportionate share of the alignment — and it's the written form of Module 6's playback, running one level up the stack.

Conclusion #

This module is forward-loaded for most readers: the skill you don't use much now becomes the skill that determines what resources you control as your scope grows. The load-bearing techniques are the 1-pager, the three lengths (30 seconds, 3 minutes, 30 minutes, prepared before the meeting), and the explicit ask. Build them before you need them — you will not have time to build them in the moment, and the moment tends to arrive unannounced.

Predictions #

  • Your first 1-pager will be two pages, because you can't bear to cut the context. Cut it anyway. The discipline of one page is the skill; if you can do it well in one, the module has landed.
  • The 30-second version will be the part you most resist — speaking it aloud alone feels embarrassing. Record it three or four times until it stops crashing. Automatic only comes from reps.
  • Within six months, you'll catch yourself saying "my recommendation is X — three reasons" in a conversation you didn't prepare for, and it will feel less like a technique and more like how you now think. That's the calibration moment.
  • Disagree-and-commit will be the hardest discipline here: you're trained to defend correctness. The workaround that works is doing it in writing — dissent registered in the doc, commitment visible in the follow-up note. The written form lubricates what's hard verbally.
  • You will at some point deliver a 1-pager and have it ignored — unread, the decision made informally elsewhere. That's not the technique failing; it's the separate political skill of getting documents read, which lives in relationship-building the document can't do for you. Sibling curricula cover that terrain.
  • The most consequential up-the-stack moment of your next twelve months will be one you're not currently planning for — a sudden escalation, an unexpected senior listener in a routine meeting. The three-lengths habit is what makes you ready for the meeting you didn't see coming.