AI Prompt Practical Guide That Nobody Told You About — This Is How You Should Ask to Get the Right Answers
Not just 'Fix this' but how to ask. A complete practical guide to AI prompts by situation, frequently used in everyday life.
Introduction: Why Don't You Get the Answer You Want When Asking AI?
Many people use AI. But fewer people than you'd think actually get the results they want.
Most people write like this:
"Fix this" "Write a script for me" "Summarize this"
AI does produce an answer. But something feels off. You ask again. Still not quite right. Eventually you think "AI isn't that great" and give up using it.
The problem isn't AI. It's how you ask the question.
AI only thinks as much as the information you give it. Without context, it gives generic answers. Without purpose, it writes aimlessly. Without conditions, it outputs in any format.
This article picks the most common AI usage situations in daily life and shows side-by-side what people actually do wrong versus what's right. And it provides templates you can copy and paste right now.
1. Code Fixing / Debugging
Why the Wrong Approach is a Problem
❌ "Why doesn't this work? Fix it"
AI doesn't know what "this" is. Without code, it can't analyze. Without an error message, it has to guess the cause. Without knowing your environment, it might give you the wrong solution.
When you ask like this, AI imagines code and answers based on that. If it's different from yours, the answer can't be right.
The Right Approach
✅ This Python function has a bug where it returns None in specific conditions.
[Paste code here]
Error message:
[Paste error message here]
Situation where this bug occurs:
- When input is an empty list
What I want:
1. Explanation of the bug cause
2. Fixed code (show before/after together)
3. How to avoid this mistake in the future
Key point: Code + error message + occurrence situation + desired output format. These four things are necessary for a proper answer.
Copy-Paste Template
I have a problem in this {{language/framework}} code.
[Code]
{{Paste code here}}
[Error Message]
{{Paste error message here}}
[Occurrence Situation]
{{What conditions trigger the error}}
[Environment]
- Language version: {{e.g., Python 3.11}}
- Key libraries: {{e.g., FastAPI 0.100}}
What I want:
1. Root cause analysis
2. Fixed code (with comments explaining changes)
3. Things to watch out for in this pattern
2. Code Review
Why the Wrong Approach is a Problem
❌ "Review this code"
It doesn't know what perspective to review from. Is it performance, readability, security, or everything? It doesn't know what level of developer is looking at it.
So AI picks something arbitrary and gives generic feedback. "You could improve this part" kind of obvious answers are likely to come up.
The Right Approach
✅ Review this Python FastAPI code for me.
[Code]
{{Code}}
Review perspectives:
1. Security vulnerabilities (SQL injection, missing authentication, etc.)
2. Is error handling appropriate?
3. Performance optimization opportunities
Format:
- Each item as "Issue / Severity(High/Med/Low) / Fix suggestion"
- Include code examples for fixes needed
- Don't need to mention parts that are already good
Copy-Paste Template
Review this {{language/framework}} code.
[Code]
{{Code}}
[Review Perspectives] (Keep only what applies)
- Security vulnerabilities
- Performance optimization
- Readability / Code quality
- Error handling
- Test coverage
[Format]
Each issue: Problem / Severity(High/Med/Low) / Fixed code example
Also add one-line overall evaluation at the end.
3. Writing (Blog, Article)
Why the Wrong Approach is a Problem
❌ "Write a blog post about AI"
All AI knows from this request is the topic. Who is it for? How long should it be? What tone? Should you consider SEO? Is it for people who already know about it or beginners? It knows nothing.
So you get generic, obvious writing. An "What is AI" kind of article you feel like you've seen before somewhere.
The Right Approach
✅ Write a blog post with these conditions.
Topic: How to make a chatbot with Claude API
Target reader: Developers who know Python basics but are new to AI APIs
Goal: Read this and make a working chatbot in 30 minutes
Word count: Around 1500 words
Tone: Friendly and practical. Minimize theory, focus on example code
Structure:
1. Finished screen (describe in text)
2. What you need (include pip install)
3. Step-by-step code explanation
4. Example of running results
5. Common errors and how to fix them
SEO keywords: Claude API, Python chatbot, AI API integration
Copy-Paste Template
Write a blog post with these conditions.
Topic: {{topic}}
Target reader: {{e.g., non-developer office workers / beginner developers / senior developers}}
Goal: {{What readers should gain from this article}}
Word count: {{Number of words or characters}}
Tone: {{e.g., professional / friendly / humorous}}
Structure:
- Introduction: {{How to start}}
- Main sections: {{What content should be included}}
- Conclusion: {{How to end}}
Include: {{e.g., code examples / tables / real cases}}
Avoid: {{e.g., overly technical terms / theory-heavy}}
SEO keywords: {{3 main keywords}}
4. SNS / Social Media Content
Why the Wrong Approach is a Problem
❌ "Write an Instagram post"
What account is it for? What's the content? What image goes with it? How many hashtags do you need? With nothing, AI writes a generic post for any account.
The Right Approach
✅ Write a post for my Instagram account (@_dev, AI/development content).
This post's topic:
"Good prompts make your work 10x faster"
Tone: Practical and relatable. Not too salesy.
Length: Caption around 150 characters
Hashtags: 10 relevant hashtags
End with one engagement question for followers
Emojis: Use naturally
Copy-Paste Template
Write a {{platform}} post.
Account type: {{e.g., AI/development info account / lifestyle / food blog}}
Post topic: {{Topic or message to convey}}
Tone: {{e.g., friendly / professional / humorous / motivational}}
Length: {{Character/word count or "short/medium/long"}}
Include:
- {{Number}} hashtags
- {{Emoji usage or not}}
- {{CTA: likes / comments / link clicks}}
5. YouTube / Shorts Script
Why the Wrong Approach is a Problem
❌ "Write a YouTube Shorts script"
What's the topic? How long is it? Who's it for? What's the vibe? With nothing, you get a plain script for any channel. Weak hook, murky message.
The first 3 seconds of Shorts are critical—if you don't grab viewers in that time, they swipe down. This 3-second hook is the most important part of the script, but you can't create it properly with just "write a script."
The Right Approach
✅ Write a YouTube Shorts script.
Channel: AI usage tips channel (50K subscribers, target 20-40 year old office workers)
Topic: "Ask ChatGPT like this and the answer quality increases 3x"
Length: 55 seconds (based on speaking speed, about 165 words)
Tone: Fast and practical. Friendly casual speech.
Structure:
- Hook (0~3 seconds): Shocking fact or question that makes viewers stop
- Problem statement (3~15 seconds): How most people misuse AI
- Core content (15~45 seconds): 2-3 specific tips with examples
- CTA (45~55 seconds): Encourage subscribe or comments
Additional:
- Include key words marked with [subtitle] for captions
- Give simple directions for visual elements to show on screen
Copy-Paste Template
Write a YouTube {{video type: Shorts/regular}} script.
Channel type: {{Channel topic and main viewers}}
Video topic: {{Topic}}
Length: {{Seconds or minutes}}
Tone: {{e.g., friendly / professional / humorous}}
Speech style: {{e.g., formal speech / casual / mixed}}
Structure:
- Hook (first {{seconds}}): {{How to start}}
- Main content {{seconds}}: {{Core message to convey}}
- CTA {{seconds}}: {{Desired action: subscribe / comments / likes}}
Include:
- Mark key words with [subtitle] for captions
- Mark visual instructions with (screen)
6. Email / Business Documents
Why the Wrong Approach is a Problem
❌ "Write an email. It's about scheduling a meeting"
Who are you emailing? What's the relationship? What meeting is it? Are there proposed dates? Without this, AI writes the most standard email format. Your company culture or relationship with the recipient won't be reflected at all.
The Right Approach
✅ Write a business email.
Recipient: CTO at a startup I'm contacting for the first time (been following on LinkedIn)
Goal: Request a 30-minute meeting to explore partnership possibilities
About me: AI consulting company CEO, 3 years experience
Proposed meeting dates: May 8 (Wed) 2-5 PM, or May 9 (Fri) morning
Format: English, professional but not rigid
Length: Under 150 words, concise
Must include: One sentence on why I want to meet, one sentence on what value I bring, clear CTA
Copy-Paste Template
Write in {{email / Slack message / official document}} format.
Recipient: {{Recipient title, company, relationship}}
Sender info: {{Your title, company, relevant background}}
Goal: {{What you want to achieve with this message}}
Tone: {{e.g., formal / friendly / collaborative}}
Length: {{Character/word count or "short/medium/long"}}
Must include:
- {{Item 1}}
- {{Item 2}}
Avoid: {{e.g., long greetings / unnecessary elaboration}}
7. Summarization
Why the Wrong Approach is a Problem
❌ "Summarize this" + [long text]
What's the purpose of the summary? One-line summary or itemized? What core points should be extracted? The result might be too long, or emphasizes unwanted parts, or has the wrong format.
The Right Approach
✅ Summarize this meeting notes for me.
[Text]
{{Meeting notes}}
Summary purpose: Share with team members who couldn't attend
Format:
- One-line summary (compress overall content)
- Key decisions (bullet points, max 3)
- Next action items (format: Responsible person - Task - Deadline)
- Next meeting date (if applicable)
Exclude: Chitchat, repeated content, unresolved discussions
Copy-Paste Template
Summarize this {{text type: article/meeting notes/research paper/report}}.
[Text]
{{Content}}
Summary purpose: {{e.g., presentation / team sharing / personal learning / SNS sharing}}
Format:
- {{e.g., one-line / 5 bullet points / section by section}}
- {{e.g., table format / narrative format}}
Word count: {{Character count or "as concise as possible"}}
Must include: {{Core data / conclusion / action items}}
Exclude: {{Repeated content / examples / footnotes}}
8. Translation
Why the Wrong Approach is a Problem
❌ "Translate this to English"
What's the context of the text? Is it an official document or SNS post? Will a native speaker read it or a non-native? Without this info, awkward, literal English might result.
The Right Approach
✅ Translate this Korean text to English.
[Original]
{{Text}}
Translation purpose: Business plan for US investors
Tone: Professional and persuasive
Level: Business English, natural for native speakers to read
Note: No literal translation; interpretation to match English conventions is fine
After translation: If any awkward or unnatural expressions, provide alternatives too
Copy-Paste Template
Translate this {{source language}} text to {{target language}}.
[Original]
{{Text}}
Translation purpose: {{e.g., email / SNS / official document / presentation}}
Audience: {{e.g., native speaker / non-native / expert}}
Tone: {{e.g., formal / casual / professional}}
Allow: Interpretation {{allowed/not allowed}}
Additional request: {{e.g., Suggest alternatives for awkward phrases / Show original and translation side by side}}
9. Image Analysis
Why the Wrong Approach is a Problem
❌ [Upload image] "Look at this"
It doesn't know what to look at. Do you need text from the image read? Design evaluated? Errors found? "Look at this" gives AI no direction.
The Right Approach
✅ [Upload image]
This is a screenshot of my website's main page.
Analyze from these perspectives:
1. First impression — Can someone new understand what this site does in 3 seconds?
2. Readability — Are font size, color contrast, and spacing appropriate?
3. CTA — Is the most important button noticeable?
4. Mobile — Would it look good on mobile? (considering screen size)
For each item, provide "Current state / Problem / Improvement suggestion"
Copy-Paste Template
[Attach image]
This image is {{image description: screenshot/photo/diagram}}.
Analysis perspectives:
{{What you want analyzed}}
Output format: {{By item / table / free description}}
Focus especially on: {{e.g., upper right / text section / graph}}
If anything is unclear, mark it as "unclear" rather than guessing.
10. Brainstorming / Ideation
Why the Wrong Approach is a Problem
❌ "Give me ideas"
What field? What constraints? What form of ideas? This vague request gets the most common ideas you'd find on the internet.
The Right Approach
✅ Give me 20 YouTube video content ideas.
Channel topic: Practical AI tool usage
Main viewers: 30-40 year old office workers, want to use AI at work but don't know where to start
Current subscribers: 3,000 (growth stage)
Content that performed well: "Finish ChatGPT reports in 30 minutes" (120K views)
Content that underperformed: AI concept explanation videos (avg 500 views)
Conditions:
- Include numbers in titles (e.g., "5 ways", "in 10 minutes")
- Practical, immediately usable content
- Exclude ChatGPT basics already covered widely
- About half should be Shorts-compatible ideas
Copy-Paste Template
Give me {{number}} ideas for {{topic}}.
Context:
- Goal: {{Why do you need these ideas}}
- Audience: {{Who is this for}}
- Constraints: {{Budget, time, skill level, team size, etc.}}
Reference:
- What worked: {{What was effective in the past}}
- What to avoid: {{Already tried or doesn't fit}}
Format:
- Idea title + one-line explanation
- Implementation difficulty (Easy/Medium/Hard)
11. Data Analysis / Excel
Why the Wrong Approach is a Problem
❌ "Analyze this data" + [table pasted]
What insights do you want? What decision are you making with this analysis? What format do you want results in? Without direction, AI gives generic statistical summary.
The Right Approach
✅ Analyze this monthly revenue data for me.
[Data]
{{Table}}
Analysis purpose: Decide where to focus marketing budget next month
Questions:
1. Which product category has the highest revenue?
2. Which category has highest/lowest month-over-month growth?
3. Are there any unusual patterns worth noting?
4. Which categories should we focus on next month and why?
Output format:
- Answer each question with its number
- Highlight key figures in bold
- End with 3-line action recommendations
Copy-Paste Template
Analyze this data.
[Data]
{{Data or table}}
Analysis purpose: {{What decision will you make with this}}
What I want to know:
1. {{Question 1}}
2. {{Question 2}}
3. {{Question 3}}
Output format:
- {{Table / bullet points / narrative}}
- Highlight key figures
- End with {{action recommendations / things to watch out for}}
12. Learning / Concept Understanding
Why the Wrong Approach is a Problem
❌ "What is RAG?"
It doesn't know what level of explanation. Elementary level or developer level? Just theory or real examples too? You end up with Wikipedia-level explanation.
The Right Approach
✅ Explain RAG (Retrieval-Augmented Generation) for me.
My background: Know Python well, understand ML basics,
but new to NLP and LLM
What I want:
1. One-sentence definition
2. Why it's needed (why LLM alone isn't enough)
3. Actual process (no code, diagram-style explanation)
4. 2 real-world use cases
5. Where to start learning
Use relatable analogies. Add simple explanations in parentheses for difficult terms.
Copy-Paste Template
Explain {{concept/technology/topic}} for me.
My background: {{Current level and related knowledge}}
What I want:
- Explanation depth: {{Beginner / Intermediate / Advanced}}
- Format: {{Analogy-focused / Example-focused / Diagram (text) / Code-inclusive}}
- Length: {{Short / Medium / Detailed}}
Include:
- {{e.g., Real use cases / Analogies / Limitations / Comparison with other concepts}}
Avoid:
- {{e.g., Too academic / Code examples / Advanced concepts}}
13. Interview Preparation
Why the Wrong Approach is a Problem
❌ "Tell me interview questions"
What company? What role? What stage? Without this, you get a generic list of questions for any interview. Your strengths, weaknesses, and prep level won't be considered.
The Right Approach
✅ Help me prepare for an interview.
Target company: Naver (200+ person IT company)
Target role: Backend developer (Python, FastAPI, 3 years experience)
Interview stage: 2nd technical interview (video call, 1 hour)
My strengths:
- 2 FastAPI projects
- Redis caching optimization experience
- Worked in team with code review culture
Concerns:
- Weak at algorithm coding tests
- No large-scale system design experience
Requests:
1. 10 most likely questions for this interview
2. Answer framework for each question (STAR technique, etc.)
3. How to naturally highlight my strengths
4. How to honestly but positively address my concerns
Copy-Paste Template
Help me prepare for a {{role}} interview.
Application info:
- Company: {{Company name or size/industry}}
- Role: {{Role name}}
- Experience: {{Years and main tech stack}}
- Interview stage: {{Written/1st/2nd/Final}}
My strengths: {{2-3 items}}
My concerns: {{1-2 items}}
Requests:
1. {{Number}} likely questions (tech / personality / judgment separated)
2. Answer guide for each
3. How to showcase my strengths
4. Strategy for addressing concerns
14. Meeting Notes / Summarization
Why the Wrong Approach is a Problem
❌ "Organize this transcript"
What format should it be in? Who's going to read it? What content should be emphasized? Without direction, AI might write long narrative summary or cut too much and miss important decisions.
The Right Approach
✅ Organize this meeting as meeting notes.
[Transcript or notes]
{{Content}}
Meeting type: Weekly sprint meeting
Attendees: 1 PM, 3 developers, 1 designer
Goal: Share last week's progress, decide this week's priorities
Output format:
- Meeting header (date/attendees)
- Main discussion topics (2-3 lines each)
- Decided items (specific)
- Action items (table: responsible person / task / deadline)
- Next meeting date
Exclude: Chitchat, repeated content, unresolved items
Language: Formal speech, concise
15. Presentation / Presentation Materials
Why the Wrong Approach is a Problem
❌ "Create a presentation about AI"
How many slides? Who's the audience? How long are you presenting? What's the key message? Without this, AI creates content-only structure with no story.
The Right Approach
✅ Create presentation structure and key content.
Presentation topic: Proposal to introduce AI coding tools to our team
Audience: Non-developer executives (3 people)
Presentation time: 15 minutes (Q&A 5 min separate)
Goal: Get approval for implementation (including budget)
Core message: "AI tools can improve development speed by 40%"
Structure request:
- 10-12 slides
- Each slide: title + 2-3 key points
- Where to put data/figures to convince executives
- Anticipated objections and responses
What executives care about most: Cost-benefit, security risks, learning burden
16. Writing Review / Feedback
Why the Wrong Approach is a Problem
❌ "Review this writing"
Just spelling? Rewrite tone? Check logic flow? Without direction, you don't know the review criteria.
The Right Approach
✅ Review this writing.
[Original text]
{{Text}}
Review perspectives (all that apply):
- Spelling / Grammar errors
- Awkward sentences → natural phrasing
- Logic flow breaks
Must preserve:
- My writing style and voice (don't change too much)
- Core argument and direction
Format:
- Full revised text
- Mark revisions with [Revised: original → revised] in separate list
- Brief explanation for each revision
Conclusion: Common Points of Good Prompts
I've covered 16 situations. Good prompts have something in common.
1. Provide context Who am I? Where will this be used? Who will see it? This background info helps AI give answers tailored to you.
2. State the purpose Not "summarize" but "summarize for team sharing." Purpose tells AI what to emphasize.
3. Specify output format Bullet points? Table? Narrative? Specifying format gets you exactly what you need. Without it, results differ every time.
4. Give constraints Length, what to avoid, what to include. These constraints keep AI within bounds.
5. Say what you don't want "Don't use technical jargon," "No literal translation," "Exclude chitchat." Negative conditions matter too.
You don't need a perfect prompt from the start. When results don't satisfy you, just say "what part bothered me and why" and ask again.
AI thinks as much as the information you give it. The more you tell it, the better results you get.
