Ever get that feeling when your ChatGPT prompts are a total hit or miss? One minute it’s a genius, the next it’s completely off track. Well, I just stumbled upon an absolute goldmine from a LinkedIn creator who laid out 26 science-backed principles for getting better results. 🤯
This isn’t just random advice; the original poster compiled a list of tested methods to supercharge your prompts. I was blown away by how simple yet effective some of these are!
Here are some of my favorites from the list:
⚙️ Setting the Stage for Success
The expert suggests you don’t just ask a question, you set the scene first.
- Assign a role: Tell the AI to act as an expert in a specific field.
- Define the audience: Specify who the answer is for (e.g., “Explain this to a beginner”).
- Use delimiters: Use things like
###to clearly separate instructions from context.
✍️ Giving Crystal Clear Commands
How you ask is just as important as what you ask. This AI professional shared some powerful phrases:
- Use direct commands like “Your task is…” and “You MUST…”
- Use positive language (“Do this”) instead of negative (“Don’t do that”).
- For extra weight, the creator even suggests adding, “You will be penalized if…”
💡 Next-Level Prompt Hacks
This is where it gets really interesting. I couldn’t believe some of these!
- The Tip Trick: Add “I’m going to tip $xxx for a better solution!” to your prompt. I was skeptical, but it’s a bizarrely effective psychological nudge for the AI.
- Chain-of-Thought: Start your prompt with “think step by step” to encourage more logical, reasoned-out answers.
- Ask for Questions: Tell the model, “Please ask me questions to get the info you need…” to ensure it has all the details before generating a response.
This is just a small sample of the fantastic list. The original post breaks down everything from example-driven prompting to getting the AI to match a specific writing style.
Seriously, go check out the full post from the original poster. It’s a game-changer for anyone who uses AI regularly!