Skip to content
Artificial Intelligence

How to create an effective AI prompt for tax and accounting professionals

Thomson Reuters Tax & Accounting  

· 5 minute read

Thomson Reuters Tax & Accounting  

· 5 minute read

Master the CREATE framework; a six-step prompting system that transforms vague AI interactions into precise, professional outputs tailored specifically for tax and accounting workflows.

Highlights

  • The CREATE framework helps tax and accounting professionals craft precise, effective AI prompts for better results.
  • Each CREATE element—Character, Request, Examples, Adjustments, Type of output, Extras—ensures clarity and tailored outputs.
  • Common prompting mistakes can be avoided by specifying context, detail, and output format for AI interactions.

 

For tax and accounting professionals, AI is only as useful as the instructions you give it. The CREATE prompting framework offers a structured, repeatable approach to get precise, professional-quality outputs — with fewer revisions and less back-and-forth.

Whether you are working with generative AI for content creation, machine learning tools for data analysis, or specialized AI assistants for workflow automation, the quality of your results depends largely on how well you communicate your needs. The CREATE framework ensures your prompts are comprehensive, clear, and tailored to produce the specific results your practice requires.

 

Jump to ↓

What is the CREATE framework?


Practical applications of the CREATE framework in tax and accounting


Common prompting mistakes to avoid


Moving forward with AI in your tax and accounting practice

 

What is the CREATE framework?

CREATE is an acronym that stands for:

    • Character
    • Request
    • Examples
    • Adjustments
    • Type of output
    • Extras

Each element is described below.

Character: Define the AI’s role

Start by establishing the role and expertise level you want the AI to assume. This sets the context for everything that follows.

Example: “You are an experienced tax accountant with expertise in corporate tax compliance and recent changes to tax legislation.”

This approach works whether you are using generative AI for research summaries, machine learning tools for pattern recognition, or virtual assistants for client communications.

Request: Be specific about what you need

Clearly articulate your request. Vague prompts lead to generic responses that require multiple rounds of refinement.

Instead of: “Help me with depreciation.”

Try: “Explain the differences between straight-line and accelerated depreciation methods for equipment purchased in 2024, including the tax implications of each.”

Examples: Provide context and patterns

Include relevant examples to guide the AI toward the style, format, or tone you prefer. This is particularly effective when producing client-facing materials.

Example: “Format your response similar to a client advisory memo, like this: ‘Based on recent regulatory changes, we recommend the following action items for your consideration…'”

Adjustments: Specify constraints and preferences

Define any limitations or specific requirements for the output.

Example: “Keep the explanation under 300 words, use bullet points for key takeaways, and avoid technical jargon that would be unclear to non-accounting professionals.”

Type of output: Clarify the desired format

Specify exactly how you want the information presented — whether as a summary, checklist, comparison table, or detailed analysis.

Example: “Provide your response as a three-column comparison table showing the method, benefits, and drawbacks of each depreciation approach.”

Extras: Add final instructions

Include any remaining requirements, such as regulatory sources to reference, disclaimers to incorporate, follow-up questions to address, or specific perspectives to consider.

Example: “Include references to current IRS guidelines and note any pending legislation that might affect these recommendations.”

Practical applications of the CREATE framework in tax and accounting

The CREATE framework adds value across a wide range of AI applications in professional tax practice:

    • Knowledge management: When using AI to search through tax codes or accounting standards, the framework helps you specify the exact regulatory context and output format needed for your research.
    • Content creation: Whether generating client communications or internal documentation, structured prompts produce consistent, professional outputs that align with your firm’s standards.
    • Compliance automation: For AI tools that assist with regulatory compliance, precise prompts help ensure the system understands the specific requirements and constraints of your situation.
    • Data analysis: When working with AI for financial analysis or pattern recognition, the framework helps you define the scope, methodology, and presentation format for your results.

Common prompting mistakes to avoid

Many professionals struggle with AI interactions because they treat the technology like a search engine rather than a collaborative assistant. Three common pitfalls to avoid:

    1. Being too vague. Broad requests produce broad answers.
      • Instead of: “Tell me about tax changes.”
      • Try: “Summarize the key provisions of the 2024 corporate tax updates that affect manufacturing companies with over $25 million in annual revenue.”
    2. Assuming context. AI has no knowledge of your specific situation unless you provide it. The more relevant detail you include, the more targeted the output.
    3. Neglecting output format. Failing to specify how you want information presented often results in formats that do not meet your needs — and extra time spent reformatting.

Refining your approach

The CREATE framework is not just a template for a single prompt — it is a foundation for building a systematic approach to AI interaction. As you work with different tools and use cases, you will develop an intuition for which elements matter most in a given context.

For routine tasks, the Character and Request elements often carry the most weight. For complex analysis, Examples and Type of Output become critical. For client-facing work, Adjustments and Extras help ensure the appropriate tone, completeness, and compliance considerations are reflected in the final output.

Moving forward with AI in your tax and accounting practice

The professionals who master effective prompting will have a meaningful advantage as AI continues to evolve in tax and accounting. The CREATE framework provides a reliable, adaptable structure that applies across tools and use cases — today and as new technologies emerge.

For a deeper look at implementation strategies and common obstacles, download our white paper: What are the key challenges of gen AI in accounting?

To explore AI-powered solutions designed for your practice, visit our AI-powered tax and accounting solutions resource center.

White paper

White paper

Overcoming accounting challenges with AI for efficiency and growth

Read white paper ↗

More answers