5 Advanced Prompting Techniques That’ll Instantly Improve Your Results with ChatGPT (or Any LLM)
Most people stop at "Act as a..." or “Write me a...”.
Tap a slide to expand
Most people stop at “Act as a…” or “Write me a…”.
But the real power of prompting comes when you start engineering the interaction — not just the output.
Here are 5 advanced prompting techniques I use regularly in business, marketing, and product development:
—
Let the AI Ask You Questions Before It Answers
💡 When to use:
Complex tasks, unclear goals, or when you’re not sure how to start.
📌 Prompt:
“Before giving me a final answer, ask me up to 5 clarifying questions so you can better understand my needs.”
✅ Why it works:
Most failed outputs are due to missing context. Letting the model ask saves time, improves accuracy, and aligns the direction before deep generation.
—
Prompt as a Collaboration, Not a Command
💡 When to use:
Strategy development, ideation, or planning.
📌 Prompt:
“Let’s work together on a launch strategy. I’ll share my initial idea, and you help expand or challenge it with data, frameworks, or counterpoints.”
✅ Why it works:
Framing the prompt as a collaboration unlocks richer insights, especially for multi-step problems or creative decisions.
—
Ask the AI to Act Like a Specific Role
💡 When to use:
You want expert-level responses or niche perspectives.
📌 Prompt:
“Act as a SaaS product manager with 10 years of experience. Give feedback on this pricing page and suggest improvements to increase conversion.”
✅ Pro tip:
Add “Use real-world SaaS examples from successful startups.” to make it more grounded.
—
Force Trade-offs to Reveal Priorities
💡 When to use:
Decision-making, product prioritization, or roadmap planning.
📌 Prompt:
“We can only ship 2 features in this sprint: A, B, or C. Which should we choose based on user impact, effort, and retention? Justify each decision.”
✅ Why it works:
Forces the model to reason instead of listing everything. You get clearer pros/cons and sharper prioritization.
—
Use “Critic Mode” to Get Honest Feedback
💡 When to use:
Editing content, improving pitches, validating strategy.
📌 Prompt:
“Be brutally honest. You’re a VC reviewing my pitch deck. What would make you say no?”
✅ Why it works:
Switching the tone unlocks a different behavior — less helpful, more skeptical. Perfect for testing robustness.
—
💬 Want better answers from ChatGPT?
Don’t just give better prompts — give it better instructions on how to think.
That’s the difference between “good” and “game-changing.”
—
I share practical tips about business, marketing and AI
🔔 Follow me to learn more!
♻️ Re-post this to help others!
🔖 Save this for future reference!
💬 DM me for collaboration!
#AI #ChatGPT #LLMs #PromptEngineering #ArtificialIntelligence #GenAI
Enjoyed this? Subscribe for more.
Practical insights on AI, growth, and independent learning. No spam.
More in Tech & Startup
Zuckerberg: Better to Overspend on AI Than Miss It
I actually respect Mark Zuckerberg for his gut as a technopreneur and innovator.
How to Push AI Output Quality Using the Generator-Evaluator Pattern
It is called "Harness Design for Long-Running Application Development" and it reads like a technical deep-dive on building front-end applications with AI. Mo...
A common question I get from my Foundations of Claude Code workshop learners:
"Is the website we built SEO and GEO optimised?"
**GenAI Pitfalls**
ChatGPT has recently encountered various outages. These outages took down our AI services and disrupted business operations, both for us and our clients.
Does AI have empathy? I asked Claude a simple product question.
The answer surprised me.
How I Balance Business and Parenting as an Entrepreneur
Last December, as I squeezed out some time to help my kids with their study, I thought:
Zuckerberg: Better to Overspend on AI Than Miss It
I actually respect Mark Zuckerberg for his gut as a technopreneur and innovator.
**GenAI Pitfalls**
ChatGPT has recently encountered various outages. These outages took down our AI services and disrupted business operations, both for us and our clients.
Does AI have empathy? I asked Claude a simple product question.
The answer surprised me.
How to Push AI Output Quality Using the Generator-Evaluator Pattern
It is called "Harness Design for Long-Running Application Development" and it reads like a technical deep-dive on building front-end applications with AI. Mo...
A common question I get from my Foundations of Claude Code workshop learners:
"Is the website we built SEO and GEO optimised?"
How I Balance Business and Parenting as an Entrepreneur
Last December, as I squeezed out some time to help my kids with their study, I thought: