Director, Learning Design & Development| PMIAsheville, NC, United States
With Generative AI, iteratively refining and optimizing prompts can lead to better AI-generated results. This may involve adjusting the specificity or clarity of the prompt to increase relevance and accuracy of results.
What examples do you have of how improving a prompt drastically changed the output quality? What specific changes did you make that led to the improvement?
Absolutely, refining a prompt can drastically change the output quality in Generative AI. Think of it like giving better instructions to a very fast, very smart assistant: the more context, structure, and clarity you provide, the more relevant and useful the result. Saving Changes...
In my experience with GenAI, refining a prompt has often made the difference between a vague, surface-level response and a precise, actionable solution—especially in a fast-paced project environment. Early on, I noticed that general prompts produced generic results, which weren’t always useful for tasks like drafting stakeholder communications, summarizing technical reports, or creating project timelines. But by refining the prompt—adding specific context, defining the audience, and clearly stating the desired format or goal—the output became significantly more relevant and usable. For example, when coordinating preconstruction documents, simply asking for “a summary” yielded too broad a result. But when I asked, “Can you summarize this scope of work for a utility-scale solar project to present to engineering leads?” the result was tailored and ready to use. That shift taught me that prompt engineering isn’t just about asking questions—it’s about guiding the AI with clarity and purpose, which directly improves quality and saves time. Saving Changes...
Just like humans can interpret instructions different if not enough context or detail is given, the same is true when using AI. Seeing the AI's initial answers helps show its strengths and weaknesses, allowing us to take that into consideration when designing future prompts. Saving Changes...
Julija AtanasovaPhD(c)| University of TorontoToronto, Ontario, Canada
In my experience, refining prompts significantly improves clarity, relevance, and precision, often transforming an average output into exactly what is needed. Saving Changes...
Sarah FlanaginProject Manager and Business Analyst| Indiana Wesleyan UniversityMarion, In, United States
Jun 21, 2024 7:28 AM
Replying to Sergio Luis Conte
...
There are framewoks to create prompt. This is part of the Prompt Desing discipline. Those that gave me and the initiatives where I was included are:R-T-F (Role-Task-Format), T-A-G (Task, action, goal), B-A-B (Before, after, bridge), C-A-R-E (context, action, result, example), R-I-S-E (role, input, steps, expectations).
Appreciate the other prompt options to get started. Thanks! Saving Changes...
Ahmed HassanConsultant civil engineer| Architect and Engineering consultantDubai, Egypt
Jun 21, 2024 7:28 AM
Replying to Sergio Luis Conte
...
There are framewoks to create prompt. This is part of the Prompt Desing discipline. Those that gave me and the initiatives where I was included are:R-T-F (Role-Task-Format), T-A-G (Task, action, goal), B-A-B (Before, after, bridge), C-A-R-E (context, action, result, example), R-I-S-E (role, input, steps, expectations).
Thanks to sharing AI to improve my quality and inspection on my side
RTf the good way to input for ai Saving Changes...
Piotr KarwotProject Manager| DXC.TechnologyCieszyn, Poland
context, examples, specific questions and clarifications help extract better responses from AI tool. Saving Changes...
Continuously refining the prompt with constraints have helped me. For example, i have instances where the LLM alters API endpoints address using something i have used in the past, but adding constraint such as do not change keeps the LLM in tune with my desires Saving Changes...
Being specific, clear, and following a pattern are not enough. It is also important to refine the prompts due to political aspects. AI avoids sharing answers that could lead people politically and financially. Therefore, refining the prompt is important to have proper answers. Saving Changes...
Anonymous
Refining the prompt makes a big difference. At first, I used general prompts and got vague answers. But when I started adding more detail — like the goal, tone, and audience — the responses became much clearer and more useful. For example, asking for a formal summary for senior managers gave me a much better result than just saying “summarize this.” The more specific the prompt, the better the output. Saving Changes...
PMI Team Member
