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?
ROHIT SEAM Global Head of Indirect R&D Sourcing and Procurement| Applied Materials Inc.Santa Clara , CA, 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).
Thanks, Sergio for sharing these frameworks. Will definitely try these in my prompts and share more feedback. Saving Changes...
It is essential to apply an iterative approach when using AI, including continuous prompt refinement, rephrasing, and the provision of clear examples and multiple scenarios to achieve the desired outcomes. While various AI models may generate results based on their underlying training and prior patterns, the project manager remains fully accountable for the final deliverables and decisions. As such, the PM should leverage multiple models or solution approaches, critically evaluate and compare the outputs, and validate them against project requirements, constraints, and success criteria. This comparative and iterative process enables informed decision-making, reduces risk, and ensures alignment with organizational objectives, consistent with PMI’s emphasis on professional judgment, governance, and accountability. Saving Changes...
SAMHAR SAMOON Consultant| AECOM Arabia Ltd.Sri Lanka Sri Lanka, 5, Sri Lanka
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.
Saving Changes...
Anonymous
A bad prompt gives you a lazy guess; a good prompt gives you a pro solution.
It's the difference between asking for "advice" and giving clear rules, a role, and a goal.
Refining your prompt turns the AI from a clueless intern into your smartest employee.
I’ve seen significant improvements in output quality by moving from vague requests to structured prompts that clearly define role, goal, constraints, and output format. For example, asking an AI to “analyze project risks” produced generic results, but refining the prompt to specify the project context, decision purpose, risk criteria, and a tabular output with mitigation actions drastically improved relevance and usability. In practice, small changes like adding success criteria, assumptions, and an explicit format often make the difference between a generic answer and a decision-ready output.
Saving Changes...
Anonymous
Thanks for all your insights. As a student currently working on my Project Management course, knowing all of these help me a lot to understand and appreciate the use of AI in PM.
Saving Changes...
Anonymous
BEtter qwuestions lead to better answers
Saving Changes...
Anonymous
Better input = better output
Saving Changes...
Betty Ann HamiltonAudit & Assurance Client CoordinatorPHILADELPHIA, PA, United States
Jun 21, 2024 10:50 AM
Replying to Laura Lazzerini
...
I think it is important to give the context and also to refine, asking for a different output in case that the first one is not completely suitable to our purpose or to the outcome that we were looking for. I think that consistency and preseverance in looking for the result, is crucial as well.
I believe effective prompting starts with providing clear context and objectives, and then iterating as needed. Consistency and perseverance in guiding the process are critical to achieving high-quality, actionable results. Saving Changes...
Key ways prompt refinement has changed output quality:
Contextual Specification & Persona Adoption: Moving from a vague command to a specific, persona-driven prompt drastically improves relevance.
Structured Output & Constraints: Defining explicit formats—such as tables, markdown, bullet points, or JSON—transforms messy text into usable data.
Iterative Refinement (Agile Approach): The process is not one-and-done. Refining prompts based on initial outputs is like an agile sprint, where instructions are tweaked until the AI aligns with the exact user requirements.
Role-Playing and Examples: Using tools like "few-shot prompting" (providing examples) allows the model to understand the desired tone and depth, often resulting in a more polished, nuanced, and accurate result
PMI Team Member
