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?
Refining a prompt can completely transform the output—vague prompts often yield generic or off-target responses, while adding context, constraints, or examples produces more accurate and tailored results. For example, shifting from “write about climate change” to “summarize the economic impacts of climate change on coastal cities in under 200 words” changes the output from broad and unfocused to precise, relevant, and actionable. Saving Changes...
Francis TauroProject Lead| Tata Consultancy ServicesPune, Maharashtra, India
I used to Gen AI very generically before getting into the knwledge of prompt engineering. From very vague answers and responses to a very useful, constructive and quality outputs has been the results of refining prompts Saving Changes...
Hasan NazmulPMO Lead| Nitto DigitalDhaka, Bangladesh
I noticed the biggest improvement while drafting my SOP for a Master’s in IT Project Management. My first prompts were broad (“write me an SOP”), so the output was generic. Things changed when I provided more context — my 16+ years in ICT across South Asia, Agile/Scrum facilitation, my startup, and future return goals. I even shared two prior SOPs from peers as reference, and asked the AI to align tone and structure while cutting irrelevant accolades (like Huawei awards) and emphasizing my current work with Nitto. Iterative tweaks such as “shorter, crisper sentences” or “focus more on my startup than past titles” completely transformed the draft into something sharp, authentic, and usable. For me, it proved that better prompts — with clarity, context, and examples — directly translate into higher-quality AI output.
...
1 reply by Shane Shahin Rad
Aug 19, 2025 2:28 PM
Shane Shahin Rad
...
By reviewing prompt results and focusing more on area of interest one can make sure the correct prompt results are obtained from latest models.
Saving Changes...
Veronica FordProject Manager| Verger Development SolutionsHouston, Tx, 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).
This is very useful as I look for more and more ways to get a more in-depth data analysis from my AI. Thank you for sharing. Saving Changes...
Delphine CooleyProgram Manager| MRM // McMannDetroit, Mi, United States
Refining prompts has enabled me to effectively assess and provide overviews on contracts. Saving Changes...
Shane Shahin RadProject Manager| Silicon ServicesGreater Los Angeles and Greater Orange County CA, United States
Aug 18, 2025 3:28 AM
Replying to Hasan Nazmul
...
I noticed the biggest improvement while drafting my SOP for a Master’s in IT Project Management. My first prompts were broad (“write me an SOP”), so the output was generic. Things changed when I provided more context — my 16+ years in ICT across South Asia, Agile/Scrum facilitation, my startup, and future return goals. I even shared two prior SOPs from peers as reference, and asked the AI to align tone and structure while cutting irrelevant accolades (like Huawei awards) and emphasizing my current work with Nitto. Iterative tweaks such as “shorter, crisper sentences” or “focus more on my startup than past titles” completely transformed the draft into something sharp, authentic, and usable. For me, it proved that better prompts — with clarity, context, and examples — directly translate into higher-quality AI output.
By reviewing prompt results and focusing more on area of interest one can make sure the correct prompt results are obtained from latest models. Saving Changes...
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 Sergio for share with us framework to create prompt. I belive that they give us a reference to elaborate and create our own prompt. I´ll test them. Saving Changes...
Anonymous
Absolutly, when you refine the question, you can get better results, when you tailoring the information you want to get, the output will be aligned with the real context. Saving Changes...
Anonymous
Absolutly, when you refine the question, you can get better results, when you tailoring the information you want to get, the output will be aligned with the real context. Saving Changes...
PMI Team Member
