🎯 Data Prompts: Teaching AI to Work With Your Information
The Magic Kitchen Analogy 🍳
Imagine you have a magic kitchen assistant who can do anything with ingredients—but only if you give clear instructions. Data prompts are like those instructions. You tell the assistant exactly what to do with the “ingredients” (your data), and it transforms them into something useful!
🌍 Translation Prompts: The Language Bridge
What Is It?
Think of translation prompts like having a friendly translator friend who speaks every language in the world. You give them words in one language, and they give you the same meaning in another.
The Story
Little Maya found a letter from her pen pal in Japan. She couldn’t read Japanese! So she asked her AI helper:
“Translate this to English: こんにちは、お元気ですか？”
And like magic, she got: “Hello, how are you?”
Simple Example
Translate the following text from Spanish to English:
"Buenos días, ¿cómo estás?"
AI Response: “Good morning, how are you?”
Real-Life Uses
- 📧 Reading emails from international friends
- 🛒 Understanding product descriptions from foreign websites
- 📚 Learning new languages by seeing translations
Pro Tips
- Always specify source and target languages
- Give context if words have multiple meanings
- Ask for formal or casual tone when needed
🏷️ Classification Prompts: The Sorting Hat
What Is It?
Remember the Sorting Hat from Harry Potter? Classification prompts work the same way! You give the AI something, and it puts it in the right category.
The Story
Tommy runs a toy store. Every day, hundreds of customer messages arrive. Some are complaints, some are questions, some are compliments. It would take forever to read each one! So Tommy asks AI:
“Classify this message: ‘I love the new teddy bear! Best purchase ever!’ Categories: Complaint, Question, Praise, Other”
AI says: “Praise” ✨
Simple Example
Classify the following movie review as
Positive, Negative, or Neutral:
"The movie was okay. Nothing special,
but not terrible either."
AI Response: “Neutral”
Categories You Can Create
| Type | Examples |
|---|---|
| Sentiment | Happy, Sad, Angry, Neutral |
| Priority | Urgent, Normal, Low |
| Topic | Sports, Tech, Food, Travel |
| Intent | Buy, Browse, Complain, Ask |
Pro Tips
- List ALL possible categories clearly
- Categories should not overlap
- Give examples if categories are unusual
🔍 Extraction Prompts: The Detective
What Is It?
Imagine you’re a detective looking for clues in a big pile of papers. Extraction prompts help AI find and pull out specific pieces of information from text.
The Story
Sarah received a long email about a party. She just wanted to know: When? Where? What to bring? Instead of reading the whole thing, she asked:
“Extract the date, location, and items to bring from this email…”
AI found:
- 📅 Date: December 15th, 7 PM
- 📍 Location: 123 Happy Street
- 🎁 Bring: A dessert to share
Simple Example
Extract all person names and their job titles
from this text:
"John Smith is the CEO. Maria Garcia works
as the Marketing Director. Tom Lee is an
intern in Sales."
AI Response:
- John Smith → CEO
- Maria Garcia → Marketing Director
- Tom Lee → Intern (Sales)
What You Can Extract
- 📧 Email addresses
- 📞 Phone numbers
- 💰 Prices and amounts
- 📅 Dates and times
- 🏢 Company names
- 👤 People names
Pro Tips
- Be specific about what you want extracted
- Ask for a structured format (list, table, JSON)
- Tell AI what to do if information is missing
🔄 Transformation Prompts: The Shape-Shifter
What Is It?
Like a caterpillar becoming a butterfly, transformation prompts change data from one form to another while keeping the same meaning.
The Story
Jake had a boring spreadsheet of customer data. His boss wanted a nice, readable report. Jake asked AI:
“Transform this CSV data into a formatted table with headers…”
The messy data became a beautiful, organized table!
Simple Example
Before (messy):
name:john,age:25,city:NYC
name:lisa,age:30,city:LA
Prompt:
Transform this data into a readable table format.
After (clean):
| Name | Age | City |
|---|---|---|
| John | 25 | NYC |
| Lisa | 30 | LA |
Common Transformations
- 📝 Bullet points → Paragraph
- 📊 Raw data → Formatted table
- 🔤 UPPERCASE → lowercase
- 📋 Long text → Summary
- 🎯 Technical → Simple language
Pro Tips
- Show the before and describe the after
- Specify the exact output format you want
- Mention if data should be sorted or filtered
❓ Q&A Prompts: The Wise Owl
What Is It?
Like asking a wise owl who has read every book, Q&A prompts let you ask questions about specific information and get precise answers.
The Story
Emma had a 50-page manual for her new camera. She just wanted to know one thing: How do I take a night photo? She gave the manual to AI and asked:
“Based on this manual, how do I take photos in low light conditions?”
AI read the whole manual and gave her just the steps she needed!
Simple Example
Based on the following text, answer the question.
Text: "The Eiffel Tower is located in Paris,
France. It was built in 1889 and is 330
meters tall."
Question: How tall is the Eiffel Tower?
AI Response: “The Eiffel Tower is 330 meters tall.”
Types of Q&A
| Type | Example Question |
|---|---|
| Factual | “What is the capital?” |
| Explanatory | “Why does this happen?” |
| Comparison | “Which is bigger?” |
| Procedural | “How do I do this?” |
Pro Tips
- Provide the context (the information to search)
- Ask one question at a time for best results
- Specify if you want short or detailed answers
📊 Analytical Prompts: The Detective + Calculator
What Is It?
Analytical prompts are like having a super-smart friend who can look at information, find patterns, spot problems, and give you insights.
The Story
Mr. Chen owned a bakery. He had sales data for the whole year but didn’t know which products to focus on. He asked AI:
“Analyze this sales data. Which products sell best? Any seasonal patterns? What should I stock more of?”
AI discovered that cupcakes sell 3x more on Fridays, and bread sales drop in summer!
Simple Example
Analyze this feedback data and identify
the top 3 customer concerns:
"Shipping is slow" - 45 mentions
"Product quality great" - 100 mentions
"Website confusing" - 38 mentions
"Love the packaging" - 60 mentions
"Prices too high" - 55 mentions
AI Response:
- 🚚 Slow shipping (45 mentions) - Most urgent
- 💰 High prices (55 mentions) - Cost concern
- 🌐 Confusing website (38 mentions) - UX issue
What You Can Analyze
- 📈 Trends over time
- 🔍 Patterns in data
- ⚠️ Problems or anomalies
- 📊 Comparisons between groups
- 🎯 Recommendations based on data
Pro Tips
- Clearly state what you want to analyze
- Specify the type of insights you need
- Provide enough data for meaningful analysis
🗺️ The Big Picture
graph LR A["📄 Your Data"] --> B{What do you need?} B --> C["🌍 Translation"] B --> D["🏷️ Classification"] B --> E["🔍 Extraction"] B --> F["🔄 Transformation"] B --> G["❓ Q&A"] B --> H["📊 Analysis"] C --> I["Same meaning, new language"] D --> J["Put into categories"] E --> K["Find specific info"] F --> L["Change the format"] G --> M["Answer questions"] H --> N["Discover insights"]
🎓 Quick Comparison
| Prompt Type | What It Does | Best For |
|---|---|---|
| Translation | Changes language | International content |
| Classification | Sorts into groups | Organizing large data |
| Extraction | Pulls out details | Finding specific info |
| Transformation | Changes format | Data cleanup |
| Q&A | Answers questions | Quick lookups |
| Analytical | Finds insights | Understanding patterns |
🚀 You’ve Got This!
Data prompts are your secret weapon for working with information. Whether you need to translate, sort, find, transform, question, or analyze—there’s a prompt for that!
Remember: Be clear, be specific, and tell AI exactly what you want. Just like giving instructions to a helpful friend, the better your instructions, the better the results!
🎉 Now go forth and prompt like a pro!