Great AI results start with clear prompts. Whether generating formulas, creating content in AI fields, or analyzing documents, specific instructions produce better outputs. This guide teaches you to write prompts that get accurate, useful results from Baserow’s AI features.
This guide covers prompt writing for:
Clear, specific prompts get better AI results. Include what you want, reference your data with {Field Name}, and specify output format. Test on a few rows before generating for your whole table.
AI models respond to the instructions you provide—your “prompt.” The quality of AI-generated content depends heavily on how well you communicate your intent. A vague prompt produces vague results; a detailed, structured prompt produces precisely what you need.
Every good prompt needs four parts:
{Field Name}Example:
Classify this review as Positive, Neutral, or Negative.
Review: {Customer Review}
Return only one word.
Classify {Input Field} as [Option1], [Option2], or [Option3].
Return only one of these options.
This pattern works well for support ticket routing, sentiment analysis, and lead scoring.
Summarize {Input Field} in [X] sentences focusing on [key aspects].
This pattern is effective for condensing meeting notes, customer feedback, or long documents while maintaining key information.
Extract from {Input Field}: Item 1, Item 2, Item 3.
If any field is missing, return "Not found"
This pattern helps pull specific data from invoices, resumes, and forms.
Write a [content type] for [audience].
Input: {Field Name}
Tone: [style]
Length: [constraint]
This pattern is ideal for creating product descriptions, email responses, and social media posts.
✅ Do:
{Customer Name}, {Order Total}❌ Don’t:
✅ Do:
❌ Don’t:
✅ Do:
❌ Don’t:
| Setting | When to use | Examples |
|---|---|---|
| 0 - 0.3 | You need consistent, factual outputs | Classification, data extraction, summaries |
| 0.4 - 0.6 | You need balanced outputs | General business writing, Q&A |
| 0.7 - 1.0 | You need creative, varied outputs | Marketing copy, product descriptions |
❌ “Analyze this data”
✅ “Identify the top 3 sales trends from {Sales Data} in Q4 2024”
❌ “Classify this: {Text}”
✅ “Classify this support ticket as Bug, Feature, or Question: {Ticket Description}”
❌ “Extract invoice total from {PDF}”
✅ “Extract invoice total from {Invoice PDF}. If missing, return ‘Unable to process’”
❌ “Analyze sentiment, extract topics, classify category, and generate response”
✅ Create separate AI fields for each task
❌ “Calculate 10% of the total column”
✅ “Calculate 10% of {Order Total}”
What you want: Flag projects that are overdue or due within 7 days
Prompt:
If {Due Date} is in the past, show 'Overdue'.
If it's within 7 days, show 'Due Soon'.
Otherwise, show 'On Track'.
What you want: Apply different discount percentages based on order quantity
Prompt:
If {Quantity} is over 100, reduce {Price} by 20%.
If {Quantity} is over 50, reduce by 10%.
Otherwise, use full {Price}.
What you want: Combine first and last name with proper spacing
Prompt:
Combine {First Name} and {Last Name} with a space between them.
What you want: Show number of days remaining until an event date
Prompt:
Calculate the number of days between today and {Event Date}.
What you want: Show how far actual spending is from budget as a percentage
Prompt:
Calculate the percentage difference between {Budget} and {Actual Spend}.
Round to 2 decimals.
Why are my outputs inconsistent?
Lower the temperature to 0.1–0.3 and add explicit constraints to improve consistency.
Why does the AI ignore parts of my prompt?
Simplify the prompt and place the most important instructions first.
Why is the AI making up information?
Add: “Only use information from Field Name. If not available, return ‘Not found’.”
Why is the output in the wrong format?
Provide a clear example: “Format as: Label: Value”
Why does it work for some rows but fail for others?
Handle empty fields explicitly: “If Field is blank, return ‘N/A’.”
Need help? Visit the Baserow community or contact support.