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?
Improving prompts by adding clear context, defined roles, specific structure, and explicit constraints transforms vague AI outputs into focused, relevant, and actionable results.
Improving prompts by adding clear context, defined roles, specific structure, and explicit constraints transforms vague AI outputs into focused, relevant, and actionable results.
Prompt engineering is clearly a path forward. Let me explain: we start with basic prompts and record them in a digital notebook. We then add more components and analyze the different outputs. By adding formulas and comparing outputs for different types of tasks, we can identify which formulas work best for which tasks. Finally, by engineering and breaking down complex tasks into smaller ones and adjusting the communication within the prompt model, we can arrive at a "standard" prompt that ensures a minimum level of satisfaction and deliverable quality. Maintaining this history allows us to have the best-performing prompts readily available, which we can use whenever needed and improve upon. It's fantastic!
Saving Changes...
Javier GamaFacilities Engineer| GeoParkBogota, Cundinamarca, Colombia
Im new using AI as a project manager. I have realized that the more clear I am and the more details I give to the model, the better the ansers. Saving Changes...
Kehinde IsaProject Manager / Regional Production Specialist| Construction, Appliance, Electrical, Lighting and ControlTexas, TX, United States
Refining a prompt can significantly improve AI output by making it more accurate, relevant, and tailored to your needs. When you add clear goals, context, and specific instructions, the response becomes more detailed, structured, and useful instead of generic. Even small changes—like specifying format or audience—can turn a basic answer into a high-quality, actionable result. Saving Changes...
In my experience, refining prompts can significantly improve output quality. For example, asking “Explain BIM” gave a generic answer. Changing it to “Explain BIM for an architect working on healthcare projects, focusing on MEP coordination at IFC stage in bullet points” made the response far more relevant. The improvement came from adding context and format. Similarly, “Write an email about delays” produced a vague response. Specifying tone, situation, and intent—like requesting a revised timeline without sounding confrontational—resulted in a much more practical email. For complex tasks, breaking them into smaller parts (scope, milestones, risks) also improved clarity and usability. Overall, the biggest impact came from adding context, clarity, tone, and structure. Saving Changes...
Anonymous
Refining the prompt can lead to minor improvements or substantial changes depending on the specific prompt and refinements.
The output was very broad, generic, and not actionable.
After refinement, I changed it to something structured:
“You are a senior project manager. Analyze risks for a software implementation project in a banking environment. Categorize risks by technical, operational, and regulatory. For each risk provide probability, impact, and mitigation actions in a table.”
This change introduced role, context, constraints, and format. The result shifted from vague text to a structured, decision-ready output.
In practice, moving from a general request to a structured prompt consistently improves relevance, accuracy, and usability of the results.
Refining and optimizing prompt give more specificity and improved results to solve the problem and meet the desired targets, goals and objectives Saving Changes...
PMI Team Member
