In the book A More Beautiful Question, Warren Berger states that questioning is a powerful, yet underutilized, tool for sparking innovation, solving problems, and driving meaningful change in both our professional and personal lives.
Most people think they know how to prompt. However, probably only 5% of them are right. The rest are exactly like I was—just one year ago: They throw a half-baked thought into ChatGPT, as if the thing could read minds.
In the previous article, Prompt Engineering — And how to get really good at it; fast!, I introduced you to the concept of AI Prompting Frameworks.
Today we are going to practice.
Today I want you to learn that 50% of the value of AI comes from knowing how to ask. And the other 50% comes from practicing the same prompts over and over until they become part of how you think.
This article is an invitation. Not to read, but to do.
Here you will find ten prompts drawn from sources like:
- MIT’s Effective Prompts for AI
- Atlassian’s Ultimate Guide to Writing Effective Prompts
- Talaera’s communication frameworks
- and FounderPath’s Top 400 AI Prompts
They are useful across roles, industries, and tools.
Each one is now a copy-paste prompt you can run immediately.
CRISP: A Beginner AI Prompting Framework
- C — Context: What you want to achieve.
- R — Role: Who the model should act as.
- I — Input: What you are giving it.
- S — Steps: How you want it to think.
- P — Presentation: How you want the answer formatted.
MIT emphasizes that effective prompts provide explicit roles, constraints, and goals, and CRISP is exactly that.
An Advanced Framework: The Iterative Loop
- Set goal
- Ask for clarifying questions
- Provide constraints
- Request a draft
- Critique it (“switch to reviewer mode”)
- Produce an improved version
- Iterate until done
Atlassian notes that this loop is one of the highest-ROI patterns for complex work.
Now let’s get to the ten prompts.
1. Rewrite for Tone and Audience
Great when you need to talk up instead of sideways.
Copy-paste prompt:
“Act as a senior communications manager in a tech company. I’m going to paste a Slack update that I originally wrote for my team. Rewrite it as a concise, non-defensive status update for C-level executives. Keep it under 150 words, make it calm and factual, and highlight:
- Current status
- Main risk
- What we’re doing about it
Preserve the meaning but improve the tone and clarity. Here is the text: [paste Slack message here].”
“Best communication prompts” lists this tone-shifting pattern as one of the most broadly useful skills for modern professionals.
— Ragan’s
Prompt break down
Let’s break down the prompt above into its CRISP parts. Once this is done you’ll gain clarity and will be able to spot these yourself.
- Context: I’m going to paste a Slack update that I originally wrote for my team. Rewrite it as a concise, non-defensive status update for C-level executives. Keep it under 150 words…
- Role: Act as a senior communications manager.
- Input: The slack message you want re-written
- Steps:
- Rewrite the message for C-level executives
- Keep it under 150 words
- Calm and factual…
- Presentation: Preserve the meaning but improve the tone and clarity.
2. Turn Chaos Into Structure
Perfect for the “brain dump” after a long day.
Copy-paste prompt:
“I’m going to paste messy notes from a quarterly business review meeting. Turn them into a clear document with the following sections:
- Key decisions
- Action items with owners and deadlines
- Open questions
- Risks and dependencies
- A brief executive summary in 5 bullet points
Do not invent new information; only reorganize what is already there.”
Communication and productivity guides (like Talaera’s prompt lists) consistently show this pattern as a top time-saver for professionals.
Notice the not-so-subtle hint at the end of the prompt: “Do not invent new information.” When you want to create repeatable outcomes, you need to think not only about what you want your LLM to do, but also what you do not want it to do.
3. Clarify and Scope Any Task
Use this when the task feels fuzzy or political.
Copy-paste prompt:
“Act as a product strategist. I’m trying to decide whether our B2B SaaS company should build an internal analytics dashboard from scratch or adopt an off-the-shelf analytics tool for our customers. Before you answer, ask me up to 5 clarifying questions that will help define the objective, constraints, assumptions, and success criteria. Do not attempt an answer until you have asked the questions and I have responded.”
Clarifying-first patterns like this are repeatedly highlighted in effective prompt guides as one of the fastest ways to improve answer quality.
4. Plan a Project or Initiative
Turn “we should do X” into a plan you can share.
Copy-paste prompt:
“You are a project manager in a SaaS company. Turn the following goal into a concrete project brief. Include:
- Project objectives
- Assumptions
- Key milestones
- Major risks and mitigations
- A simple timeline for the next 12 weeks
If any critical information is missing, ask me targeted questions before finalizing the brief. Goal: ‘Launch a small paid pilot of our new AI feature with 5 design-partner customers this quarter.’”
FounderPath’s business prompt collections show project-planning prompts like this among the most common in real-world use.
5. Break Down Complex Work
When the thing in front of you feels too big to even start.
Copy-paste prompt:
“Decompose the following initiative into a practical, step-by-step execution plan. Group steps into phases, estimate rough effort for each step using S (small), M (medium), or L (large), and highlight which steps require collaboration with design, engineering, marketing, or leadership. Initiative: ‘Redesign our product onboarding flow to improve activation and reduce time-to-value for new users.’”
MIT’s work on prompting emphasizes decomposition as one of the most effective ways to improve reasoning quality in AI systems.
6. Analyze and Summarize Information
Turn raw signal into something you can act on.
Copy-paste prompt:
“Act as a customer insights analyst. I will paste a collection of ~50 short customer feedback snippets about our product. Analyze them and provide:
- A 5-bullet executive summary
- The top 5 recurring themes or pain points you see
- 3–5 follow-up questions or analyses that would help us deepen our understanding or prioritize improvements
Focus on patterns, not individual one-off comments.”
Business research guides (like New Horizons’ AI prompt lists) position this as a core pattern for using AI in analysis.
7. Generate Options, Then Critique Them
Because good strategy is not “first idea wins.” Because hope is not a strategy.
This prompt is particularly useful. Using ChatGPT or Gemini to poke holes into one’s own assumptions is a great way to reveal biases, risks, and opportunities.
Copy-paste prompt:
“Act first as a marketing strategist, then as a critical reviewer.
Step 1: Generate 7 possible email subject lines for a webinar invite about ‘How mid-market SaaS companies can use AI to reduce churn.’ The audience is VP-level leaders in product and customer success.
Step 2: Switch roles to a skeptical reviewer and, in a table, evaluate each subject line with pros, cons, and a score from 1 to 10 for likely open rate.
Step 3: Based on your review, propose 2 improved ‘best of’ subject lines that combine the strengths of the top options.”
Atlassian and other guides recommend role-switching patterns like this to force deeper critical thinking.
8. Turn AI Into a Personal Tutor
Your “always on” senior mentor.
Copy-paste prompt:
“Act as a senior product manager and learning coach. I am currently a mid-level PM who has mostly worked on feature delivery. I want to grow into a strategic product lead who can own roadmap, discovery, stakeholder alignment, and data-driven decisions within the next 6 months.
First, ask me 3–5 questions to assess my current skill level in strategy, discovery, stakeholder management, and data-driven decision making. Then, based on my answers, design a 6-month learning plan with monthly themes, weekly practice exercises, and 2–3 concrete deliverables I should ship (for example: a product brief, a discovery report, a strategy memo).”
Prompt communities like PromptGenius frequently showcase tutoring prompts as one of the highest-impact uses of AI.
9. Decision Support and Trade-Off Analysis
When “it depends” is not good enough.
Copy-paste prompt:
“Help me decide between three analytics tools for our SaaS product: Tool A, Tool B, and Tool C. Our context: we are a mid-size B2B SaaS company with a small data team, a product-led growth motion, and limited engineering capacity for maintenance.
- Build a comparison table for A, B, and C with criteria such as implementation complexity, cost, flexibility, reporting capabilities, integration effort, vendor lock-in risk, and time-to-value.
- Based on the table, provide a clear recommendation and explain your reasoning.
- List the top 5 pieces of additional data (about our team, users, or the tools) that would most likely change your recommendation.”
Decision-support patterns like this match what Atlassian and others call “structured trade-off analysis prompts.”
10. Continuous Workflow Copilot
This is where AI stops being a search box and becomes a colleague.
Copy-paste prompt:
“You are my workflow copilot for the next 45 minutes. The project is: ‘Draft and refine a 2-page strategy memo for launching our new AI assistant inside the product.’
Your responsibilities:
- Ask me what I have already done and what I’m stuck on
- Suggest the single next concrete step I should take
- Offer to either draft, review, or improve any artifact I’m working on (bullet points, outline, memo, slides)
- Keep each response short so we can iterate quickly
- Periodically check if we are still aligned with the original 2-page memo goal”
This “copilot” pattern is very close to what TTMS describes as AI becoming “the new operating system for knowledge work.”
Your Turn
I want you to get something out of this—reading prompts is not the point.
Using the prompts is.
- Pick one real task from your day: a decision, a messy document, a project, a difficult email.
- Choose two prompts from above.
- Paste them.
- Run them.
- Do it again, and again.
That is how you actually get better at this. That is how you achieve AI prompting mastery—by doing, not by reading.
Until next time, team.