Honest Take — Module 9: Salary & Offer Negotiation — Direct Attack on the Root Cause #

This is the module the curriculum exists for, and I'm going to be more direct here than anywhere else, because for most readers the time horizon is not abstract. You have an offer window coming — the interview loop in progress, the promotion cycle opening, the recruiter thread you haven't answered — and the dollar delta between the version of you that walks into that window having done this module's work and the version that defaults to the historical pattern is, conservatively, $10,000 to$ 25,000 per year. Because bases compound forward, that single conversation is worth $50,000 to$ 150,000 over the following five years. The math is not complicated. The math is the entire reason this module is sized the way it is, and why the Live-Stakes Track exists: this module is meant to be run against a real offer, not rehearsed in the abstract.

Here is what will surprise you: you already know what to do. McKenzie's salary essay is the canonical text and you have probably absorbed its moves third-hand even if you've never read it. Don't give the first number. Anchor high when forced, with research behind the number. Treat counters as data, not threats. Negotiate the entire package. Never accept on the spot. Always have something to walk to. These are not secrets; they are not even contested. The asymmetric failure is the gap between what you know at 11 AM reading the essay at your desk and what you do at 4 PM when a friendly recruiter asks "so, what are your salary expectations?" The body chemistry between those two moments is the whole problem. The protocol fits in one paragraph — deflect the expectations question toward their band; if forced, give a researched anchor 15-25% above your gut-comfortable number; if they balk, ask for their band and let them move first; take time, never accept live, counter on the full surface (signing bonus, equity terms, vacation, start date, remote flexibility). The reason engineers don't run it for whole careers is not ignorance. It's that under live pressure the protocol evaporates and the mouth says "yeah, somewhere around X" — with X chosen by the nervous system, which is running a much older program.

So the real work of this module is somatic, and nobody writes about this part. When the question lands you will feel a specific cluster: tightening in the throat, a slight upward pitch, hot pressure to fill the silence, the urge to pad the number with "ish" and "around" and "but I'm flexible." That cluster is your body trying to defuse a conversation it has registered as confrontation, and your historical pattern is to obey it. The protocol works only if you can tolerate the cluster without obeying it — and tolerance is built by reps, not insight. Concretely: say your researched anchor out loud, alone, into a recorder — "my expectation for this role is in the range of X to Y" — twenty times, until it lands flat. If it isn't flat at twenty, do twenty more. The flatness is the deliverable, because the flatness is what the other side hears, and what they hear shapes what they offer.

Then instrument the feedback loop: within five minutes of the real call ending, record yourself narrating what was said, what you said, what you wish you'd said, and where the cluster showed up. Engineers build feedback loops for code and refuse to build them for their own behavior. This one costs nothing and teaches more than any book in the module.

One more correction, about the market itself. Your gut anchor is calibrated to a market that no longer exists — the market you entered years ago, at a seniority you no longer have, possibly in a geography that no longer bounds you. If you have a genuine specialty intersection — the legacy-stack expert at an AI company, the engineer with the deep open-source record, the specialist whose profile the hiring pipeline rarely sees — the scarcity is leverage the recruiter has already priced in even while hoping you haven't.

And remember what the unclaimed delta actually is: the hiring committee has room in the band, the recruiter expects a counter, and when you don't ask, the difference between what they would have paid and what you accepted doesn't go to charity. It goes to margin. That money was budgeted for you. The reason to take it isn't greed — it's that the futures you're funding, the people who depend on you and the goals you carry that are bigger than a career, are paid for out of exactly this delta.

Conclusion #

The protocol is one paragraph; the body chemistry is the actual opponent. Rehearse the anchor out loud until it lands flat, refuse the first number, treat counters as data, never accept live, negotiate the full package, and record the post-mortem the same day. Then run it against your next real offer window — not a roleplay — because that's the only place the habit can actually form. The dollar delta from one well-run conversation is the curriculum's entire ROI, collected in a single afternoon.

Predictions #

The first ten reps of saying your anchor aloud will produce the full body cluster. By rep twenty it will dampen; by thirty the number will land flat. The flatness is what the recruiter will hear.
Your next initial offer will arrive above your historical default, and your nervous system will register it as "too good to risk countering." It is not. It is their opening position. Counter anyway.
You will forget at least one intangible on your first counter — signing bonus, equity terms, vacation. Ask for it on the second counter; most candidates never make a second counter at all, and the second one almost always lands something.
The same-day post-mortem will reveal exactly one moment where you softened your number unnecessarily. Mark it. It will not recur on the following call.
You will feel guilty for "asking too much" within 24 hours of countering. The guilt is the old pattern resurfacing; it passes within a week, and the money compounds for a decade.
The delta between doing this work and not doing it will be at least $10,000/year on your next offer, conservatively — and you'll only be able to verify that if you wrote your pre-call numbers down, so write them down.
Your second negotiation after this module will be noticeably easier than the first, and by the fifth the protocol will be the default rather than the rehearsed exception. The module produces a habit, not an event.