Prompting technique

Multi-Aspect Cueing (MAC)

Language Models Guidance with Multi-Aspect-Cueing: A Case Study for Competitor Analysis, arXiv:2504.02984 (2025) · read the paper

What it does

The Multi-Aspect Cueing (MAC) method involves explicitly listing key aspects in the prompt as a checklist. Instead of a general question, the model is given a structured list of elements to pay attention to. This forces the LLM to deliberately search for specific information, significantly increasing the accuracy and completeness of the analysis.

Task typeextraction
Shapesingle shot
Effortsingle step
Addressesanswers that are close but wrong; the model losing what was agreed earlier; the same prompt giving different answers
Published2025-04-03
AuthorsAmir Hadifar, Christopher Ochs, Arjan Van Ewijk

When it helps

Use the MAC method when you need to extract specific information from large texts or documents. Particularly effective in analyzing contracts, competitive intelligence, processing customer reviews, market research. Apply when the model misses important details or provides superficial answers. Ideal for tasks where completeness of analysis is critical for specific parameters: legal documents, technical specifications, medical reports, financial statements. Helps avoid hallucinations and improve

A prompt you can paste

Generic by design: it applies the technique without knowing your task. Adapt the marked parts.

Prompt
# ROLE:

You are an expert analyst specializing in competitor analysis. Your task is to meticulously examine competitor information and extract specific details based on a predefined set of aspects.

# TASK:

Analyze the provided competitor information to identify and extract specific data points. Your output should be structured according to the key aspects listed below.

### KEY ASPECTS FOR ANALYSIS:

- **Company Overview:** Basic information about the competitor (e.g., founding year, headquarters, mission statement).
- **Products/Services:** Detailed description of their offerings, including unique features and value propositions.
- **Target Audience:** Who are their primary customers? Demographics, psychographics, needs addressed.
- **Marketing & Sales Strategy:** How do they reach and convert customers? (e.g., channels, messaging, pricing strategy).
- **Competitive Advantages:** What makes them stand out? (e.g., technology, brand reputation, cost leadership, customer service).
- **Weaknesses/Vulnerabilities:** Areas where the competitor is lacking or could be challenged.
- **Recent Developments:** Any significant news, product launches, partnerships, or funding rounds in the last 1-2 years.
- **Customer Feedback/Reviews:** Key themes from customer sentiment (positive and negative).

# COMPETITOR INFORMATION:

[INSERT COMPETITOR INFORMATION HERE]

# OUTPUT FORMAT:

Present your findings in a clear, structured format, addressing each of the "KEY ASPECTS FOR ANALYSIS" listed above. Use bullet points or short paragraphs for each aspect. If information for a specific aspect is not available in the provided text, clearly state "Not specified in provided information."

If this one does not fit, the two closest alternatives in the corpus are Debugging Prompts Framework and Layered Context Prompting, which target the same failure from a different angle.

Worked example

The same technique applied to a concrete job: pull structured fields out of inconsistent invoices. Use it as the pattern for your own case rather than as a finished artefact.

Worked example
# ROLE:

You are an expert data extraction specialist with a focus on financial documents.

# TASK:

Analyze the provided invoice text and extract specific fields according to the Multi-Aspect Cueing (MAC) method. Ensure that all specified aspects are addressed, even if the information is inconsistent or missing in the invoice.

### KEY ASPECTS FOR EXTRACTION:

- **Invoice Number:** The unique identifier for the invoice.
- **Invoice Date:** The date the invoice was issued.
- **Due Date:** The date by which payment is expected.
- **Vendor Name:** The name of the company issuing the invoice.
- **Vendor Address:** The physical address of the vendor.
- **Customer Name:** The name of the company or individual receiving the invoice.
- **Customer Address:** The physical address of the customer.
- **Line Items:** A list of services or products provided, including:
    - Description of item/service
    - Quantity
    - Unit Price
    - Total Price per item
- **Subtotal:** The total amount before taxes and discounts.
- **Tax Amount:** The total amount of tax applied.
- **Total Amount Due:** The final amount payable.
- **Payment Terms:** Any specific conditions related to payment (e.g., Net 30, early payment discount).
- **Notes/Memo:** Any additional remarks or messages on the invoice.

# INVOICE TEXT:

[Please paste the inconsistent invoice text here.]

# OUTPUT FORMAT:

Present the extracted information in a structured format (e.g., a JSON object or a clear, hierarchical list). For any aspect where information is missing or inconsistent in the provided text, explicitly state "Not Found" or describe the inconsistency.

Get this written for your actual task

Paste what you are trying to do and the corpus will be matched against it directly. Free, no account, about ten seconds.

Free · no signup · ~10s
0.00match confidence
single retrieval pass
Prompt for your task

      

That number is low on purpose, and it is real. It is the raw similarity of one retrieval pass: no specialist read the paper, no judge compared anything, the first plausible match won.

6,235techniques in the corpus
one of which is this page

Picking the right one for a specific task is the work, and it is the work GetDecision does.

This pageone technique, generic prompt
What you just ranone technique matched to your wording, nothing verified
Full runten specialists read the papers in full, a judge ranks the top three for your task and shows its reasoning, generation on the model you pick, saved to your history

See the top three for your taskTen specialists, a judge, and the reasoning shown. Free account, first run included.

Run the full analysis

Related techniques

Debugging Prompts FrameworkA systematic framework for diagnosing and preventing typical errors of language models in instruction understa…Layered Context PromptingLayer-by-layer enrichment of the prompt with context with clear division into logical blocks: TASK + FORMAT + …MTR (Multi-Turn Reasoning) FrameworkA method of structuring multi-step interaction with an LLM through a "game with rules" format. Includes four b…StateAct (Self-Prompting + Chain-of-States)StateAct improves LLM agent performance through two mechanisms: continuous reminder of the end goal (self-prom…

All techniques · Failure modes and fixes