Are you feeling that none of your AI outputs are underwhelming, inconsistent, or just plain wrong?
The problem usually isn't the tool. It's the prompt.
We tend to mostly talk to AI, expecting for it to read our minds with few words.
But success comes when have a real conversation with it and are guiding each other through every step.
When you master proper prompting techniques, you can increase your AI output quality by 150% or more.
You'll save hours of back-and-forth corrections, get exactly what you need on the first try, and finally unlock AI's true potential for your work.
5 essential AI prompts
This guide breaks down five simple frameworks that will instantly improve the way you work with AI:
APE — Get clarity with Action, Purpose, and Expectation.
RACE — Improve results with Role, Action, Context, Example.
RISE — Structure your thinking with Role, Input, Steps, Expectation.
CARE — Set clear expectations with Context, Action, Result, Example.
ROSES — Think strategically with Role, Objective, Scenario, Expectation, Steps.
What is a prompt?
A prompt is the instruction you give to an AI system to accomplish a specific task.
While you might ask a coworker, "Can you help me with our marketing strategy?", an AI needs more structured information to deliver quality results.
It's like the difference between telling someone to "make dinner" versus providing a detailed recipe with ingredients, cooking methods, and timing.
Good prompting is about clarity, not complexity.
Poor vs. Effective Prompts
Poor prompt example: "Help me with marketing."
Effective prompt example: "Create a 3-month content marketing strategy for our B2B SaaS company targeting CTOs at mid-sized tech companies. Focus on establishing thought leadership in AI automation. Include content pillars, distribution channels, and success metrics. Format as a actionable roadmap with weekly milestones."
An effective prompt tells the AI exactly what to create, for whom, with what focus, and in what format. This specificity leads to dramatically improved results.
Why learning to prompt matters
We used to say that content is king. But, there’s been a fundamental shift—now, it's context that rules. AI performs exponentially better when you provide rich context about you, your business, and your specific needs.
Based on our training of over 500 people in AI, more than 90% of users copy prompts from LinkedIn "AI gurus" without understanding the underlying principles. This leads to:
Generic outputs that don't match their specific needs
Frustration when the AI doesn't deliver expected results
Missed opportunities to leverage AI's full capabilities
Wasted time on multiple revision cycles
When you master effective prompting, you see remarkable results:
Teams save 8-10 hours per week,
they achieve $600-800 in weekly productivity improvements per participant,
and transform your employees into more confident, strategic-thinking workers.
Good prompting equals better AI adoption and results. It's that simple.
The 5 key elements of any good prompt
Every effective prompt contains five core components:
1. Task
What you want the AI to accomplish. Be specific about the deliverable. Example: "Create a competitive analysis report"
2. Instructions
Step-by-step directions for how the AI should perform the task. Example: "Use bullet points, include pricing comparisons, and highlight key differentiators"
3. Context
Background information that helps the AI understand your situation. Example: "We're a startup competing against established players in the project management space"
4. Parameters
Constraints like tone, length, style, or creativity level. Example: "Keep it professional, limit to 2 pages, focus on actionable insights"
5. Input
Data, examples, or resources that inform the response. Example: "Here are our last three competitor analysis reports as reference"
Note: Including examples of previous work you've done can increase output quality by at least 150%. If you've created something similar before, include it as a reference in your prompt.
APE: Action, Purpose, Expectation
Best for: Simple, straightforward tasks that need clear direction and quick results.
When to use APE:
Content creation tasks
Simple strategy development
Quick analysis requests
Daily productivity tasks
Template:
Action: Define the specific job or activity.
Purpose: Explain the intention or goal.
Expectation: State the desired outcome.
Example: Building a content marketing strategy
Action: Develop a content marketing strategy for our new line of sustainable linen blend clothing.
Purpose: We want to make people excited about our clothes and increase brand awareness with our target audience—people interested in sustainable clothing that is timeless, resistant, and durable.
Expectation: The strategy should engage our audience, create strong brand recall, and drive a 25% increase in pre-orders compared to our last product line.
Tips:
Keep it concise, but specific.
Use APE with faster AI models like GPT-5 or Claude Sonnet for quick turnaround.
Perfect for tasks that don't require deep strategic reasoning.
Great for kickstarting brainstorming or campaigns.
Great starting point for beginners learning to prompt.
RACE: Role, Action, Context, Example
Best for: Strategic planning, complex business scenarios, and tasks requiring specialized expertise.
When to use RACE:
Strategic business planning
Market analysis and research
Complex problem-solving scenarios
Tasks requiring specific professional expertise
Model recommendation: Use GPT-5 Thinking for strategic tasks requiring reasoning and strategic thinking.
Template:
Role: Who should the AI act as?
Action: What needs to be done?
Context: What’s the situation?
Example: What does a good output look like?
Example: Market entry strategy for a SaaS company
Role: Act as a senior business strategist with expertise in SaaS market expansion.
Action: Develop a comprehensive market entry strategy for international expansion into three European markets.
Context: We are a mid-size German SaaS company offering project management software. We lead our domestic market and want to expand into France, Netherlands, and Spain. Our customer base includes SMBs with 50-250 employees, and we have strong data privacy compliance capabilities.
Example: The strategy should include market prioritization criteria, competitive landscape analysis, regulatory compliance requirements, localization needs, partnership opportunities, and an 18-month systematic entry timeline with specific milestones and success metrics.
RISE: Role, Input, Steps, Expectation
Best for: Step-by-step processes, content strategy development, and detailed workflows.
When to use RISE:
Project planning and management
Instructional content creation
Process development and optimization
Data-driven strategy building
Structure:
Role: Who is the AI?
Input: What data or audience info should it use?
Steps: How should it build the output?
Expectation: What is the final deliverable?
Example: Content strategy for a specific target audience
Role: Act as a content strategist with expertise in B2B marketing and thought leadership.
Input: Our target audience consists of CTOs and IT directors at fintech companies with 100-500 employees. They struggle with legacy system integration and are actively researching automation solutions. [Attach: audience research document, competitor content analysis]
Steps: Provide a step-by-step content strategy including: 1) Content pillar identification based on audience pain points, 2) Channel selection and optimization approach, 3) 90-day content calendar framework, 4) Performance measurement and optimization strategy.
Expectation: The strategy should generate 3 qualified leads per month within 90 days and establish our CEO as a thought leader in fintech automation.
Model selection tips:
Use GPT-5 for straightforward content planning
Switch to GPT-5 Thinking when incorporating complex data analysis
Consider using Projects in ChatGPT for document organization and ongoing iteration
8. CARE: Context, Action, Result, Example
Best for: Goal-oriented tasks with specific, measurable outcomes where situational awareness is crucial.
CARE puts context first, making it ideal when the current situation heavily influences the solution approach.
When to use CARE:
Goal-oriented business initiatives
Customer-facing strategy development
Product launch planning
Outcome-focused problem solving
Structure:
Context: What’s the situation?
Action: What should the AI do?
Result: What’s the goal?
Example: What does success look like?
Example: Building a 3-month go-to-market strategy
Context: We are a B2B software company launching a new integration feature that connects with popular accounting platforms like QuickBooks and Xero. Our existing 500+ customers have been requesting this feature for 18 months, and our main competitors don't offer similar integrations.
Action: Create a comprehensive 3-month go-to-market strategy for this new integration feature launch.
Result: Achieve 40% adoption among existing customers and attract 50 new customers specifically interested in this integration capability, generating $150K in additional revenue.
Example: The strategy should include customer communication sequences, feature demonstration approaches, pricing considerations, training materials, and success metrics. Reference how Slack approached their workflow automation rollout—gradual release, extensive customer education, and integration showcases.
Tip: CARE works great when you're under pressure to produce specific results.
9. ROSES: Role, Objective, Scenario, Expectation, Steps
Best for: Complex problem-solving scenarios with multiple stakeholders, variables, and interdependencies.
When to use ROSES:
Leadership and people operations challenges
Complex organizational problem-solving
Multi-stakeholder initiative planning
Situations requiring systems thinking
Model recommendation: Use GPT-5 Thinking for complex reasoning and comprehensive analysis.
Structure:
Role: Who should the AI be?
Objective: What’s the high-level goal?
Scenario: What’s happening right now?
Expectation: What do you want as a result?
Steps: What steps should the AI follow?
Example: Improving team productivity and prioritization
Role: Act as an experienced project manager and organizational psychologist with expertise in team productivity and burnout prevention.
Objective: Improve our marketing team's productivity and job satisfaction while reducing burnout and improving work quality.
Scenario: We have an 8-person marketing team overwhelmed with competing priorities. Team members work overtime regularly, miss deadlines frequently, and express frustration about unclear priorities. Recent survey results show declining job satisfaction (down 30% in 6 months) and increased stress levels. Leadership frequently changes priorities mid-project, and there's no systematic approach to workload management.
Expectation: Develop practical prioritization frameworks, improved communication processes, and workload management systems that can be implemented within 30 days and show measurable improvements in 60 days.
Steps: Create a comprehensive solution including: 1) Priority scoring methodology that leadership will actually use, 2) Weekly planning framework with built-in buffer time, 3) Communication protocols for priority changes, 4) Workload monitoring approach with early warning systems, 5) Team feedback mechanisms to prevent future issues.
Advanced tips and model selection
Choosing the right AI model
GPT-5: Best for quick content creation, analysis, and daily productivity tasks. Fast response times make it ideal for iterative work and multiple prompt testing.
GPT-5 Thinking: Use for strategic planning, complex reasoning, market research, and any task requiring deep analysis.
Claude Sonnet: Use for content creation, analysis, and most day-to-day AI assistant needs.
Claude Opus: Best for complex reasoning and analysis, advanced problem-solving, and nuanced understanding.
Custom instructions and memory features
Maximize effectiveness by setting up:
Custom instructions:
"Cut the fluff and provide actionable insights"
"Match my direct communication style"
"Always include specific examples and implementation steps"
"Format responses with clear headers and bullet points"
Memory features:
Let your AI tool remember your role, company, and frequent tasks
Build context over time instead of re-explaining background
Use "save this to memory" for important preferences and information
ChatGPT Projects vs GPTs: When to use each
Use Projects when:
Organizing multiple related conversations over time
Working with documents and files regularly
Switching between different AI models for the same topic
Collaborating on ongoing initiatives with file uploads
Use GPTs when:
Building specialized tools for specific use cases
Using custom actions and API integrations
Sharing AI assistants with team members
Wanting pre-configured prompting for repeated tasks
Testing and iteration strategy
Don't rely on a single approach. Test your prompts across:
Different models: Compare GPT-5 speed vs 5-Thinking depth
Various platforms: Test ChatGPT, Claude, and Gemini for different strengths
Multiple iterations: Edit previous messages to try different approaches
Pro tip: You can edit any message in a conversation and regenerate responses without starting over. This saves time when refining prompts.
Bringing it all together
You now have five powerful frameworks for different scenarios:
APE for simple, focused tasks requiring quick clarity.
RACE for strategic business challenges needing expertise.
RISE for systematic processes and step-by-step planning.
CARE for goal-oriented initiatives with specific outcomes.
ROSES for complex problem-solving with multiple variables.
As you practice each framework, you'll naturally internalize the five core elements (task, instructions, context, parameters, input) and start prompting intuitively.
The ultimate goal is to automatically think,
"How can AI make this better, faster, and more effective?" for every task you encounter.
You'll stop copying prompts from others and start creating your own based on your specific needs and context.
Most importantly, view AI as a collaborative partner that amplifies your capabilities rather than replaces them. The best results come from combining human creativity and judgment with AI's processing power and knowledge.
The future belongs to those who can effectively communicate with AI—and now you have the roadmap to do exactly that.
Want to get even better, faster? Join our AI-First Mindset Training or subscribe to our newsletter for weekly AI tips, tools, and prompts that actually work.
Tim Cakir
CEO & Founder
