In a recent LinkedIn post, Andrew Bolis tackles the common frustration surrounding the quality of responses generated by AI tools like ChatGPT. Rather than blaming the technology itself, Bolis argues that the key to achieving expert-level results lies in the art of prompt engineering.
According to Bolis, the perceived shortcomings of AI output are often a direct consequence of poorly constructed prompts. He emphasizes that users must provide clear, detailed instructions to guide the AI effectively. As Andrew Bolis notes:
“Stop blaming ChatGPT for low-quality responses. Bad output results from poorly written prompts.”
Bolis then proceeds to offer a practical solution: a set of nine distinct prompt frameworks designed to elicit more precise and valuable responses from AI models.
Unpacking Bolis’s Prompt Frameworks for Enhanced AI Interaction
Andrew Bolis’s post details several structured approaches to prompt writing, each tailored to different output objectives. These frameworks aim to provide users with a systematic way to communicate their needs to ChatGPT, thereby improving the relevance and quality of the generated content.
The R-A-C-E and R-I-S-E Frameworks
Bolis introduces the R-A-C-E framework, which stands for Role, Action, Context, and Expectation. This structure encourages users to first define the persona ChatGPT should adopt (e.g., an editor or coach), then specify the actions it should perform, provide necessary background information, and finally, clearly state the desired outcome. Similarly, the R-I-S-E framework (Role, Identify, Steps, Expectation) focuses on defining a role, identifying a core issue, outlining actionable steps, and setting clear expectations for the final output.
Leveraging STAR, SOAP, and CLEAR for Structured Output
Further elaborating on his structured approach, Bolis presents the S-T-A-R (Situation, Task, Action, Result) method, often used in behavioral interviewing, adapted for AI prompting. This framework helps in providing a narrative context. The S.O.A.P. (Subject, Objective, Action, Plan) method is suggested for outlining problems and solutions, while the C.L.E.A.R. (Context, Learn, Evaluate, Action, Review) framework provides a comprehensive structure for tasks involving learning and evaluation.
“Try a framework today to quickly improve your results.”
As Andrew Bolis highlights, these frameworks are not just theoretical constructs but practical tools for immediate improvement.
Advanced Frameworks for Comprehensive Prompting
Bolis also shares more complex frameworks like P.A.S.T.O.R. (Problem, Amplify, Story, Transformation, Offer, Response), which guides users through a problem-solving narrative, and F.A.B. (Features, Advantages, Benefits), a classic sales technique useful for product descriptions or value propositions. The universal 5-W-1-H (Who, What, When, Where, Why, How) method is presented as a means to ensure all critical aspects of a query are covered. Finally, the G.R.O.W. (Goal, Reality, Options, Will) model, commonly used in coaching, is adapted to help users define objectives and actionable steps.
The Importance of Prompt Quality
Andrew Bolis’s central thesis is that the efficacy of AI tools is directly proportional to the quality of the input prompts. By adopting structured frameworks, users can move beyond generic queries and achieve more sophisticated, expert-level results. Bolis encourages readers to experiment with these methods to enhance their own AI interactions.
“Here are 9 prompt frameworks for expert-level results:”
The post concludes with a call to action, inviting readers to save the post for future reference and follow Bolis for more AI-related insights.
📝 About This Content
This article is based on insights shared by Andrew Bolis on LinkedIn.
📅 Originally posted on November 23, 2025 | View original post on LinkedIn →