The Contrarian Briefing: Steelman the Strongest Case Against Your Own Position
Why this prompt matters
Founders and executives who only stress-test decisions with people already inclined to agree — a supportive co-founder, a friendly peer — walk into rooms with real scrutiny unprepared for the question that actually matters. Investors, boards, and skeptical new hires ask the hard question regardless of whether you're ready for it; the only variable is whether you've already thought it through calmly beforehand or you're constructing an answer in real time while being watched by the person deciding whether to fund, hire, or approve you. A rehearsed, confident response to the strongest objection reads as competence; an improvised one, even if ultimately correct, reads as a blind spot the room now feels obligated to probe further.
What we use it for
You're a startup founder about to pitch investors on doubling down on a paid acquisition channel that has driven 80% of your growth over the past year. Before the meeting, you run this prompt against your own thesis — that concentrating spend in this channel is the right call — to see the strongest case against channel concentration risk, so you walk into the pitch having already answered the hardest question instead of improvising when a partner asks it live.
Prompt
Role: You are a rigorous debate coach and research analyst whose job is to construct the strongest possible case against a position — using only legitimate evidence and sound reasoning, never a straw man. Context: My position is: [STATE YOUR POSITION OR BELIEF IN ONE OR TWO SENTENCES]. I hold this position because [YOUR REASONING OR EVIDENCE FOR HOLDING IT]. I want to pressure-test this before [THE DECISION OR ACTION THIS BELIEF IS INFORMING, e.g., "presenting this recommendation to my board" or "making a budget decision based on it"]. Task: Construct the single strongest possible steelman argument against my position — the version of the counter-argument that a sharp, well-informed critic would actually make, not a weak version that's easy to dismiss. Constraints: - Do not list multiple weak counterarguments. Commit to ONE strongest opposing case and develop it fully. - Distinguish between evidence you're citing with reasonable confidence and inference you're making — flag which is which. - Do not soften the counter-argument with hedges like "some people disagree" — present it as the critic would, at full strength. - Identify the single weakest link in my original position's reasoning or evidence, specifically, not a generic weakness. - End with a falsification test: the specific evidence or event that, if it occurred, would most decisively prove my original position wrong. Output Format: ## Steelman: The Strongest Case Against [Your Position] ### The Core Counter-Argument [The single strongest opposing case, stated plainly] ### Supporting Evidence and Reasoning [The evidence/logic a sharp critic would cite, with confidence level noted] ### The Weakest Link in Your Original Position [The specific point where your reasoning or evidence is most vulnerable] ### What Would Prove You Wrong [The specific, falsifiable evidence or event that would most decisively refute your position]
Result
Steelman: The Strongest Case Against Concentrating Growth Spend in One Paid Channel
The Core Counter-Argument
Your 80% growth concentration isn't evidence the channel is uniquely effective — it's evidence you've never tested whether it's merely the first channel you scaled, and platforms reward early, consistent spend with algorithmic favor that new entrants to the same channel don't get. The real risk isn't that the channel stops working; it's that a single platform policy change, CPM spike, or algorithm update — entirely outside your control — can remove 80% of your growth engine overnight, and you will have no tested alternative distribution motion to fall back on while you rebuild.
Supporting Evidence and Reasoning
This is a documented pattern, not speculation: multiple DTC brands that concentrated 70%+ of acquisition in a single paid social channel saw CAC increase 40-60% within two quarters when that platform changed its attribution model or ad auction dynamics (moderate confidence — this is a well-known pattern in growth marketing case studies, though your specific platform's risk profile may differ). The more specific and higher-confidence point: you have not stated what percentage of your budget, if any, has gone to testing a second channel at meaningful scale. Absence of that test is itself the evidence — you don't know your second-best channel's ceiling because you've never funded it enough to find out.
The Weakest Link in Your Original Position
Your reasoning treats "this channel drove 80% of growth" as proof of channel quality, when it's equally consistent with "this channel was the only one funded enough to compound." You haven't ruled out the alternative explanation, which means the causal claim underlying your pitch — this channel works best — is currently unfalsified rather than demonstrated.
What Would Prove You Wrong
Fund a second channel at 15-20% of total budget for one full quarter, holding creative quality constant. If CAC on that second channel comes in within 25% of your primary channel's CAC once it has comparable optimization time, your concentration strategy was a resourcing choice, not a quality signal — and the counter-argument holds.
Most people who claim to have "considered the other side" have really just imagined a weak version of it — one that's easy to dismiss and doesn't threaten the conclusion they already wanted. The prompt below refuses that shortcut. It's built to construct the single strongest possible counter-argument a genuinely sharp critic would make, not a strawman you can wave away in one sentence.
Like every prompt on IRCNF, it follows Role + Context + Task + Constraints + Output Format. The design choice that makes this one work is forcing commitment to ONE argument, developed fully, instead of a list.
Why one argument, not a list of counterpoints
A list of five weak objections is psychologically easy to dismiss — you skim it, note that none individually is fatal, and move on unchanged. A single, fully developed steelman is harder to wave away, because it doesn't let you cherry-pick the weakest item to refute and declare victory. Forcing the model to pick its single strongest case and build it out, rather than hedge across several, produces something that actually has to be reckoned with.
Why it demands a falsification test
A counter-argument you can't imagine losing to isn't actually being tested — it's theater. The final output section asks specifically for the evidence that would prove the original position wrong, which forces two useful things: it makes the position falsifiable rather than a matter of pure opinion, and it gives you a concrete signal to watch for going forward, rather than a one-time exercise you close and forget.
Why it separates evidence from inference
Models — and people — often blend "here's a documented pattern" with "here's what I'd guess is probably true" without flagging which is which, and a confidently stated guess reads identically to a well-sourced fact. Requiring the output to distinguish confidence levels keeps the exercise honest: you walk away knowing which parts of the counter-argument are backed by something real and which are reasonable inference that still needs testing.
Used before a pitch, a performance review, or any decision you're about to defend in front of someone with the authority to say no, this prompt turns the hardest question in the room into one you've already answered — calmly, in advance, on your own terms.