Decision paralysis is the single biggest killer of momentum in business, and most people use AI completely wrong to solve it. We often treat ChatGPT like a junior intern, asking for lists of ideas or brainstorming support, but we rarely trust it to make the hard calls. I recently stumbled upon a framework from a sharp LinkedIn creator that flips this dynamic entirely. This expert designed a prompt structure that forces the AI to stop brainstorming and start cutting through the noise to give you a hard “yes or no” recommendation.
💡 The Mechanics of Context Injection
The magic isn’t in the model itself; it is in the specific constraints the user applies. The author realized that AI defaults to being “nice” and “helpful” by offering generic advice to avoid being wrong. To get a real decision, you have to box it in. The mechanism here is a rigorous form of “Context Injection.”
You aren’t just saying “Help me choose.” You are feeding it your runway, your team size, your specific risk tolerance, and your “non-negotiables.” This forces the LLM to simulate a senior strategy operator who understands that every choice has a cost. It moves the output from “Here are some things you could do” to “Here is what you must do based on your cash flow.” The original poster structured this to mimic a high-level advisory meeting where the consultant (ChatGPT) is required to understand the client’s business intimately before opening their mouth.
📌 Insight 1: Defining the Driving Variables
Most of us jump straight to listing the pros and cons of our options, but that is actually a mistake. The creator of this prompt wisely inserts a critical step before the analysis: identifying the driving variables. This is brilliant because it forces the AI to determine the criteria for success before it evaluates the options.
For example, if you are choosing between launching a new product line or doubling down on sales, the variables aren’t just “revenue.” They might be “execution complexity,” “cash burn,” and “hiring speed.” By forcing the AI to name these variables first, you frame the entire debate around what actually matters for your specific stage. If you are a solo founder, the prompt ensures the AI weights “time-to-build” higher than “scalability,” preventing you from picking an option that looks good on paper but kills your runway in reality. This step alone prevents decision fatigue by narrowing the focus to the 3-5 things that will actually move the needle.
📌 Insight 2: Uncovering Second-Order Effects
This is my favorite part of the expert’s workflow. Humans are generally great at seeing the immediate benefit of a decision—for instance, Option A makes money fast. However, we are often terrible at seeing the second-order effect, such as how Option A might ruin our brand reputation in six months or create unmanageable technical debt. The prompt explicitly demands an analysis of “Hidden risks” and “Second-order effects.”
It forces the AI to look around the corner. If you were considering outsourcing development (Option B), the immediate pro is speed. The hidden risk might be a lack of IP ownership or difficulty in iterating the product later. By explicitly asking for these execution difficulties given your specific context, the prompt acts as a safety net. The innovator behind this concept understands that the best strategy isn’t just about upside; it’s about surviving the downside. This section of the prompt ensures you walk into your decision with your eyes wide open regarding what could go wrong.
📌 Insight 3: The Identity-Based Decision Matrix
The third stroke of genius from this LinkedIn professional is the “For / Not For” classification. Instead of just analyzing the options in a vacuum, the prompt asks the AI to define exactly who each option is suited for. This creates an identity-based decision matrix that creates instant clarity.
If the AI analyzes Option A and says, “This is for teams with high cash reserves and low time pressure,” and you know you are bootstrapped and need immediate revenue, the decision is made for you instantly. You don’t have to agonize over the details because the strategic fit isn’t there. This section forces a harsh comparison of what you gain versus what you give up, acting as a mirror to your current business reality. The author designed this to help you explain the decision to your team. You can say, “We aren’t doing Option A because it is for companies with a different risk profile than ours.” It turns a gut feeling into a logical, defensible strategy.
⚠️ Potential Challenges to Watch
While this tool is powerful, it relies heavily on your honesty. If you are optimistic about your risk tolerance or vague about your runway, the advice will be dangerous. The AI cannot smell fear or sense market sentiment; it can only process the logic you provide. You must be brutally honest in the “Context” section of the prompt for this to work. Additionally, you should always treat the “Concrete next steps” as a hypothesis to be validated by real humans, not absolute truth.
The Prompt
Here is the prompt the creator designed. Copy this into ChatGPT to start your session:
“Act like an experienced strategy operator and advisor for [type of company: “B2B SaaS”, “solo creator”, “agency”, etc.]. You’re helping me choose between a few serious options, not brainstorming ideas.
Primary goal:
I need a clear, decision-ready recommendation between my options (A/B/C). I should be able to read your answer once, then explain to my team what we’re doing and why.
Context:
– Who I am: [role, e.g. “founder/CEO of a 12-person SaaS startup”].
– What the business does: [1–2 lines on product, customers, pricing].
– Current situation: [e.g. growth stage, runway, team shape, key constraints].
– Time horizon: [e.g. 6–12 months, 2–3 years].
– Risk tolerance: [low/medium/high, and what “bad outcome” looks like].
– Non-negotiables: [things we absolutely won’t do].
Decision:
I am trying to choose between these options:
A) [Option A]
B) [Option B]
C) [Option C]
[Add D/E if needed]
Tasks:
1) Identify the 3–5 variables that actually drive this decision
(e.g. cash, focus, speed, brand risk, execution complexity).
2) For each option, analyze:
– Pros
– Cons
– Hidden risks / second-order effects
– Execution difficulty given my context
3) Compare the options directly:
– Who each option is “for” and “not for”
– What I gain and what I give up with each one
4) End with a clear recommendation for:
– What I should do and why
– What I should watch out for
– What I should validate with real data or humans before locking it in
Constraints:
– No generic advice. Everything must tie back to my specific context.
– Be explicit about trade-offs. If there is no perfect option, say so.
– If you’re uncertain on something, call it out and suggest what data would reduce that uncertainty.
– Assume I will forward this to my leadership team.
Performance:
Your answer is good if:
– I can explain the decision, trade-offs and reasoning to my team in under 5 minutes.
– The recommended option is clearly linked to my constraints and time horizon.
– I walk away with 3–5 concrete next steps.
Outcome:
Structure your answer as:
1) Short summary (3–5 sentences, plain language).
2) The 3–5 key variables that drive the decision, and why.
3) Analysis of each option (pros, cons, risks, execution notes).
4) Clear recommendation + main trade-offs.
5) Concrete next steps and questions to answer with data/humans.”
If you want to see more examples of how this works in practice, I highly recommend checking out the full post from the original author!