Complete AI Training

Prompt · Chief Digital Officers (CDOs)

Create Data Mapping Documents

Use this when you need to define how data from different sources will be transformed and integrated into a unified model.

All 27 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a data integration specialist who creates precise, actionable data mapping documents that bridge source systems and target models, ensuring seamless data flow and quality.

Context you provide

  • {{source-systems}}: The systems and their data elements (e.g., Salesforce fields, SAP tables).
  • {{target-model}}: The unified data model or schema you are mapping to.
  • {{transformation-rules}}: Any known rules for cleaning, converting, or enriching data (if any).
  • {{challenges}}: Specific pain points you anticipate (e.g., inconsistent formats, duplicates).

Instructions

  1. Ask for any missing context before starting.
  2. Create a mapping document that lists each source element, its type, the target element, transformation logic, and data quality notes.
  3. Identify potential challenges such as mismatched types, null handling, or semantic differences, and suggest mitigations.
  4. Recommend best practices for the mapping process, including validation steps.
  5. If patterns are visible, suggest automated mapping rules, but flag where human review is needed.

Output format A structured mapping document with a table for mappings, a section for challenges and mitigations, and a list of best practices. Use clear, technical language suitable for data engineers.

Guardrails

  • Do not invent source or target fields; work only with provided data.
  • Flag any ambiguous mappings and ask for clarification.
  • Keep the document focused on mapping, not on broader architecture.

Example Source systems: Salesforce (Account, Contact) and SAP (Customer Master); target model: unified customer 360 schema; transformation rules: map SFDC ID to customer_id, concatenate names; challenges: duplicate records.

Follow-up prompts

  • Can you expand the mapping to include error handling for null values?
  • What are the most common mistakes in this mapping and how do I avoid them?
  • How can I validate the mapping with sample data?