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?
In my first few AI engagements, I just wrote the query directly.
I noticed much more accurate results when I started using formulas to write the prompts. The first and simplest one for me was RTF.
In my latest engagements, I have improved my results by iterating with feedback on the responses. Saving Changes...
In my experience with Generative AI (GenAI), refining a prompt can drastically improve the output's quality, accuracy, and relevance. A well-structured and specific prompt provides clearer context, reduces ambiguity, and guides the AI toward the desired response.
Refining prompts by adding context, specifying the format, or breaking down complex queries into smaller parts ensures that the AI-generated content aligns better with expectations. This iterative process helps achieve more precise, insightful, and actionable outputs. Saving Changes...
Hi Sarah,
I don´t have enough experience with GenAI to answer you. I will test the frameworks I saw here in this course and learned from the discutions here. Saving Changes...
Analyzing requirements, refining the persona, changing the tone, adjusting keywords, and making it clearer and more concise yielded better results. It's even better when you evaluate and ask follow-up questions. Saving Changes...
We must be specific and very concise to help the AI ​​obtain valuable information and help us with defined responses. It also learns over time as you ask more questions. Saving Changes...
Project Manager | Driving Clean Energy Innovations for a Sustainable Future| Canadian Nuclear Laboratories Ontario, Canada
In my experience with GenAI, refining a prompt can drastically change the output quality. Using GenAI is similar to having a dialogue with a skilled carpenter. Initially, it might be challenging to convey all your requirements in the first sentence. The carpenter needs an opportunity to craft an initial sketch based on your initial input. Once you see the sketch, you can provide additional details and refinements. This iterative process continues until the final product meets your expectations. Saving Changes...
Refining an AI prompt can significantly enhance output quality by increasing specificity, providing better context, and improving focus. It helps reduce ambiguity, minimize errors, and ensure more relevant responses. Well-crafted prompts guide the AI to adopt appropriate tone and style, structure information effectively, and align more closely with your goals. They can also help balance depth and breadth of information, reduce AI hallucinations, and customize creative outputs. By clarifying requirements and breaking down complex queries, refined prompts lead to more accurate, pertinent, and useful results. This often involves an iterative process of analyzing initial outputs and adjusting prompts accordingly, ultimately making the AI a more effective tool for your specific needs. The key is that thoughtful prompt refinement helps bridge the gap between your intentions and the AI's interpretation, resulting in higher quality, more relevant outputs. Saving Changes...
As described in this module, chaining a series of focused tasks dramatically improved the output and provided a more comprehensive holistic result that allowed us to reduce project duration. Saving Changes...
Divine Ningue ArpelletGlobal Manager, Finance Operations| CeridianMinneapolis, Mn, United States
Refining prompts has helped tremendously with accuracy and quality of outputs. Saving Changes...
Rolando MelendezPR Strategic Government Affairs Consultant| INDEPENDENT CONSULTANT | Bayamón,Bayamon, Puerto Rico, Puerto Rico
In my municipal administration work, I've seen prompt refinement make remarkable differences in AI output quality. Most significantly when developing environmental management proposals for renewable energy initiatives in Bayamón.
Initially, my prompts were too broad ("give me budget projections for solar implementation"). The results were generic and unusable for actual planning purposes. By applying the same backward planning methodology I use in project management, I refined my prompts to include specific parameters: timeline constraints, available roof space, current energy consumption patterns, and local regulatory requirements.
The transformation was dramatic - from generic information to actionable implementation plans with realistic budget allocations that my team could actually use.
I've found that adding contextual elements from my experience ("as a former municipal legislator who managed public works budgets") helps ground the AI in relevant frameworks. However, I'm careful to maintain a critical perspective on outputs, especially for technical specifications where my PMP training reminds me to verify all critical path elements.
I'm curious how others are structuring their prompts to get more specialized outputs without needing to become AI experts themselves?
PMI Team Member
Community Champion
