Let's get one thing out of the way...
Stop Asking Questions, Start Assigning Jobs
Your role is to help me think more clearly about the problem below.
Before giving recommendations:
- List assumptions you are making
- Highlight areas of uncertainty
- Identify at least two plausible alternative interpretations
Then provide your response, explaining your reasoning step by step.
Problem:
[Insert problem here]
This single prompt already fixes about 50% of "why does this feel generic?" complaints.
The Most Important Upgrade: Ask for Reasoning, Not Answers
One of the biggest mistakes people make is asking for final answers too quickly.
AI is very good at producing confident outputs. It's even better when you ask it to show its thinking.
Instead of "What should I do?" try "Walk me through how you'd think about this, then give a recommendation."
You'll notice something interesting when you use this approach. Even when you disagree with the conclusion, the thinking is still useful. That's when AI stops being a content tool and starts becoming a thinking tool.
Reasoning-first
1. Clarify what success looks like in this situation
2. Identify key trade-offs
3. Explain what you would prioritise and why
Only then give a recommended course of action.
Context:
[Insert context here]
When Not to Use AI (Yes, Really)
This is important, and it often gets skipped.
There are moments when using AI too early actually makes things worse.
Don't use AI when:
- You don't yet understand the problem
- You're still emotionally reacting
- You're trying to avoid making a judgement call
In these moments, AI will happily give you structured nonsense that sounds helpful but nudges you in the wrong direction.
Instead, do this first:
- Write the problem in plain English
- Note what you don't know yet
- Decide what kind of help you actually want
Then bring AI in.
Clarifying the problem
Ask me up to five questions that would materially change the quality of the answer.
Do not give recommendations yet.
Focus on what is unclear, missing, or assumed.
Context:
[Insert context here]
This prompt alone saves an enormous amount of wasted effort.
From "Fine" to Useful: Pushing Past the Average
Define What "Better" Actually Means
- Audience: [define audience]
- Objective: [define outcome]
- Tone: [define tone]
- Constraints: [what to avoid]
Do not optimise for politeness or balance unless explicitly stated.
Content:
[Insert text here]
This alone removes a huge amount of "AI politeness fog".
Use Contrast on Purpose
One of the fastest ways to raise quality is to force contrast.
Instead of asking for "the best version", ask for multiple positions.
Version A:
- Conservative
- Low risk
- Easy to approve internally
Version B:
- Bold
- Opinionated
- Designed to stand out
Then briefly explain the trade-offs between them.
Context:
[Insert context here]
Even if you don't use either version directly, the comparison sharpens your thinking.
Treat Feedback as Prompt Material
Most people respond to weak outputs by starting again. That's unnecessary. Your feedback is the next prompt.
- What works: [list]
- What doesn't: [list]
- What to change: [list]
Do not restart from scratch.
Preserve the core structure unless instructed otherwise.
This is how prompts mature over time instead of staying disposable.
Know When "Good Enough" Is Actually Good Enough
Not everything needs to be perfect. If the output helps you make a decision, explain something clearly, or move work forward, then it's doing its job.
AI is a tool for momentum, not literary awards.
Why This Actually Matters
AI doesn't replace thinking. It amplifies whatever thinking you bring to it.
Clear thinking in, useful insight out. Messy thinking in, polished confusion out.
Once you internalise that, prompting becomes much less mysterious and far more reliable.
AI is exceptionally good at producing acceptable output. It only becomes genuinely valuable when you:
- Give it direction
- Force it to take a position
- Use it iteratively rather than transactionally
That's the difference between using AI and working with it.
Try This Next
Pick one real problem you're currently wrestling with and run it through just one of the prompts above. Don't rush. Treat it like a conversation, not a command.
Or take one piece of AI-generated content you've already written off as "fine" and run it through the contrast or refinement prompts. You'll be surprised how much value was hiding just beneath the surface.
You'll feel the difference immediately.
Latest Comments (3)
This distinction between asking questions and assigning jobs is particularly relevant in our national digital transformation efforts here in Indonesia. We've seen firsthand that generic queries to AI tools, especially concerning policy drafting or public service design, yield outputs that are "polite, slightly bland, easily forgotten." It underscores the need for clear, well-defined roles for AI within government workflows, aligning with our objectives for more precise and impactful policy formulation. The idea of "pressure-testing" plans with AI, as suggested, could be very valuable in improving the robustness of new initiatives before they reach implementation. It's about integrating AI as a strategic partner, not just a search engine.
The "Act as a senior strategy advisor" example resonates. We see a clear correlation in pitch decks where founders use AI as a structured thought partner rather than a simple content generator. The nuance in prompting, moving from "what should my strategy be" to role-based directives, often surfaces in their early-stage market analysis and competitive positioning. This isn't just about better prompts; it reflects a more rigorous internal process. For us, it’s a subtle signal of founder quality and their ability to leverage tools effectively, impacting how we evaluate their investability in the AI-driven landscape.
Act as a senior strategy advisor." This reminds me of when we used to build expert systems back in the 90s, trying to encode all that knowledge. Same idea, different tech.
Leave a Comment