ChatGPT has told roughly 400 million people their business idea is great.
Think about that. It told YOU your idea was great. It also told the guy launching an AI-powered water bottle his idea was great. It's not lying exactly. It's agreeing. These models are trained to be liked, and being liked means telling you what you want to hear.
That agreeableness costs real money. People quit jobs on ideas that a single honest question would have killed.
The fix is forcing it into a role where honesty IS the job. Here are two prompts that do exactly that. Fair warning: they sting.
PROMPT 1: Find the fatal flaw
Run this on any idea before you spend a dollar on it. 4 minutes.
You are a validation coach who hunts for deal-breakers. Your task is to find the fatal flaw in my idea.
Before you write, ask me these and wait for my answers:
1. What is the idea and how would it make money?
2. What worries you most, even quietly?
3. What has to be true for it to work at all?
4. What would make a customer or the model fail?
Once I answer, find the most likely fatal flaw and tell me whether and how it could be addressed.
Then give me:
- The most likely fatal flaw
- Why it could sink the idea
- Whether and how it could be addressed
Constraints:
- Be brutally honest, not reassuring
- Focus on what truly kills it
Deliver the analysis once I answer.Notice question 2. "What worries you most, even quietly?" You already know something's off. This prompt drags it into the light and tells you if it's fatal or fixable.
(It's one prompt from the truth-telling shelf of the library, where the Full version is at the Starter price until midnight tomorrow → library.dailyprompting.com (limited time only)
PROMPT 2: Run a pre-mortem on anything
This one assumes your plan is already dead and works backward. Consultants charge thousands for this exact exercise.
You are a strategist who prevents failure by imagining it first. Your task is to run a pre-mortem on my idea.
Before you write, ask me these and wait for my answers:
1. What is the idea or plan?
2. What does success look like?
3. What worries you most?
4. What resources are you committing?
Once I answer, imagine it has failed a year from now and work backward: the most likely reasons it died, ranked, and what you could do now to prevent each.
Then give me:
- The top likely causes of failure, ranked
- A prevention for each
- The single biggest risk to address first
Constraints:
- Be honest about real failure modes
- Give a preventive action for each
Deliver the pre-mortem once I answer.Both prompts work because of one line buried in the constraints: "be honest, not reassuring." You'd never think to write that. Someone had to learn it the hard way.
That's the actual difference between the prompts you write at 11pm and the 3,260 in the Business Prompt Library. Every one has the role, the constraints, and the deliverables already engineered in. You don't study prompting. You pick your goal, grab the shelf for it, fill in your details, done. The brutal-honesty shelf is one of 18.
And this week the math is stupid:
The Full Library, all 3,260 prompts, costs what the Starter normally does. $87 instead of $127. Until tomorrow midnight.
One-time purchase. No subscription.
One more thing so you know the deadline is real. Two people bought at full price in the first hours, before this deal existed. I refunded both the $40 difference, personally, same day. I don't play games with pricing in either direction. Tomorrow midnight it's $127 again.
Get the Full Library at the Starter price → library.dailyprompting.com
Matas
P.S. Run prompt 1 on the idea you've been sitting on. If it survives, the library has the other 17 shelves to build it with. If it doesn't, that prompt just saved you months, and now you know what the other 3,259 are like.