Acceldata
Acceldata
Search docs… ⌘K
Last updated:

Writing Effective Prompts

Prompt Structure

Basic Formula

[Context] + [Specific Request] + [Output Format]

Example

  • Context: “For the customer_orders table”
  • Request: “show me data quality issues”
  • Format: “in the last 24 hours”

Full Prompt: “For the customer_orders table, show me data quality issues in the last 24 hours.”

Components of a Good Prompt

  1. Be Specific
    • ✓ “Show failed policies for customer_orders table”
    • ✗ “Show me some policies”
  2. Provide Context
    • ✓ “As a data steward reviewing monthly compliance”
    • ✗ [No context provided]
  3. State Your Goal
    • ✓ “I need to create a report for management”
    • ✗ [Unclear purpose]
  4. Specify Output Format
    • ✓ “Provide as a table with columns for policy name, failure count, and asset”
    • ✗ [No format specified]

Context Provision

Why Context Matters

Context helps ADM:

  • Select the right tools and agents
  • Adjust tone and depth of responses
  • Apply domain-specific knowledge
  • Deliver accurate, relevant examples

Types of Context

Role Context

  • “As a data engineer maintaining production pipelines…”
  • “From a compliance officer’s perspective…”
  • “As someone new to data quality…”

Temporal Context

  • “For Q3 2025…”
  • “In the last 24 hours…”
  • “Since the last deployment…”

Scope Context

  • “For all customer-related tables…”
  • “Within the finance database…”
  • “Across all data sources…”

Background Context

  • “We recently migrated to cloud storage…”
  • “Our policy requires 99% quality…”
  • “This is for regulatory reporting…”

Specificity and Clarity

Vague vs. Specific

Vague

Specific

“Show me policies”

“Show me active data quality policies for production tables.”

“What’s wrong?”

“What data quality issues occurred in the customer database today?”

“Find tables”

“Find tables containing customer PII data.”

“Create a policy”

“Create a data quality policy to validate email formats in the user_accounts table.”

Use Concrete Terms

  • ✓ “customer_orders table” ✗ “that table we use”
  • ✓ “last 7 days” ✗ “recently”
  • ✓ “policies with < 90% success rate” ✗ “bad policies”

Avoid Ambiguity

  • Ambiguous: “Show me the data.” (Which data? Where? When?)
  • Clear: “Show me the row count for the customer_orders table for each day in the last week.”

Examples

Question Answering

  • Poor: “How’s the data?”
  • Good: “What is the current data quality score for tables in the customer database?”
  • Best: “Show me the data quality scores for all tables in the customer database, highlighting any below 95% over the last 30 days.”

Policy Creation

  • Poor: “Make a policy.”
  • Good: “Create a data quality policy for the orders table.”
  • Best: “Create a comprehensive data quality policy for the orders table that validates:
    • Order amount is positive and less than $1M
    • Order date is not in the future
    • Customer ID exists in the customer table
    • Status is one of: pending, shipped, delivered, or cancelled”

Troubleshooting

  • Poor: “Why did it fail?”
  • Good: “Why did the customer_validation policy fail?”
  • Best: “The customer_validation policy failed at 2 PM today with 450 rows. Analyze the failure pattern and suggest possible causes for the email validation rule failure.”

Analysis

  • Poor: “Tell me about problems.”
  • Good: “What data quality problems exist?”
  • Best: “Analyze data quality trends for the finance database over the last 30 days. Identify:
    1. Most frequently failing policies
    2. Tables with declining quality scores
    3. New issues that appeared this month Provide recommendations for improvement.”