Project Management

Please login or join to subscribe to this thread

In your experience with GenAI, how has refining a prompt drastically changed the output quality?

linkedin twitter facebook   Artificial Intelligence  
avatar
Sarah Philbrick
PMI Team Member
Director, Learning Design & Development| PMI Asheville, 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?
Sort By:
< 1 ... 70 71 72 73 74 75 76 77 78 79 80 ... 147 >
avatar
Anonymous
It helped absolutely! Refinement is always needed.
avatar
Wasiu Azeez Project Coordinator| TowalTech Inc. Dartmouth NS, Canada
@Sarah Philbrick acknowledged
avatar
MARIA RUIZ ARIAS Other| Oleoducto de los Llanos Orientales BOGOTA, DC, Colombia
Jun 21, 2024 10:27 AM
Replying to TAIWO POPOOLA
...
Being concise and specific helps the AI to give some valuable answers. It also learns with time as you ask further questions.
Considero clave realizar las pruebas y evaluar los resultados permite indentificar mejoras.
avatar
Sundaram Saravanan Manager - Delivery| Virtusa Consulting Services Pvt Ltd Chennai, Tamil Nadu, India
Through Iteratively refining the outcome and providing more precise requirements makes the LLM understand the requests more clearly and provide the expected result. Iterative refinement makes us understand how the LLM behaves or responds and also making the LLM understand the requirements more clearly and generate output
avatar
Wolfgang Sosa Venezuela (Bolivarian Republic of)
Adding Constraints and Desired Format.
Example. - Concise, bullet-point summary.

Now, why do they improve output?
The AI now knows exactly what kind of output is expected, preventing verbose or free-form text.
For a Project Manager, specifically focusing on how it impacts team collaboration and project adaptability": This introduced specific audience and thematic focus. Instead of a generic overview, the AI is guided to tailor the information to a project manager's perspective and highlight aspects of agile.
Ensure each point includes a brief explanation: This added a structural requirement within the format, ensuring each point isn't just a heading but provides some detail.
avatar
Alfonso Guevara Project Manager III (PM3)| JaeVeo LLC Downey, Ca, United States
In my experience, refining a prompt isn't just about making it more specific—it's about aligning it with stakeholder intent. For example, when working on a stakeholder analysis matrix automation via GenAI, my initial prompt simply requested categorization based on power/influence. The output was generic and borderline unusable. I rephrased the prompt to simulate a seasoned project manager preparing for a governance meeting: “Act as a senior PM preparing a stakeholder analysis for an executive steering committee. Categorize stakeholders using a power-interest grid and justify each placement with 1–2 strategic implications.” The result wasn’t just better—it mirrored the tone and rationale I’d expect from a senior consultant. It reminded me that great prompting is less about writing code and more about designing context. We don’t just input data—we architect the conversation.
avatar
Israel Ufomadu Whitby, Ontario, Canada
I have just recently started using AI seriously and since beginning the prompt engineering course and refining my prompts, the quality of my outputs are getting significantly better.
avatar
Anonymous
massively
avatar
Nura Habibu Babura/Babura, Jigawa/Nigeria, Nigeria
After series of Iteration, GenAI has provided better outputs in many scenarios.
avatar
Dr. Freeman Jackson, D.Sc., PMP, CISSP, CISA Tallahassee, Fl, United States

Absolutely, Sarah — refining a prompt can be the difference between a generic response and a transformative one. One example from my work involved training an AI agent to generate contract summaries. Initially, the prompt was too open-ended: "Summarize this contract." The output was vague and missed key provisions.



After iterating, I reframed it using R-T-F (Role, Task, Format): "You are a legal analyst. Summarize this contract by identifying the parties, obligations, terms, and risks in bullet-point format." The result was dramatically clearer, more relevant, and immediately actionable.



Prompt engineering isn’t just about getting better answers — it’s about communicating expectations with precision. That’s a core project management skill in a GenAI-driven world.



Looking forward to hearing how others are using structured prompt methods like R-T-F or C-A-R-E in their workflows!



— Freeman Jackson
< 1 ... 70 71 72 73 74 75 76 77 78 79 80 ... 147 >

Please login or join to reply

Content ID:
ADVERTISEMENTS

Wow! They've got the internet on computers now!

- Homer Simpson

ADVERTISEMENT

Sponsors







Vendor Events

See all Vendor Events

Newsletters

Jun 3: The Career Problem Facing New PMs
May 20: Your PM Career as a Project Life Cycle
May 13: The Big Lie After Layoffs

See all newsletters