Mastering Prompt Engineering - CREATE formula

CREATE Prompt Formula

If you’re looking to maximize your interaction with ChatGPT/Gemini or any other language model out there and extract the most useful information from it, you’ve come to the right place. I’ve discovered an effective framework with the help of PMI of course “The CREATE framework”, which stands for Character, Request, Examples, Adjustments&Constraints, Types of Output, Evaluation&Steps.

Image Credits: ChatGPT 

CREATE formula is a more elaborate and customized version of the RTFS formula ( refer to my blog on RTFS ) providing a more comprehensive framework where we can give specific instructions to the language model to provide an output close to expectations.

Here are the details of create formula followed by some examples

Component	Description	Example from the Prompt
Character	Define the role or identity the AI needs to assume in order to complete the task effectively.	"You are a meticulous project manager, specializing in managing renovations and construction within healthcare facilities."
Request	Clearly state the task or objective that the AI is expected to perform.	"Analyze and identify potential risks in transforming a software architecture to optimize the traffic flow."
Examples	Share relevant examples, case studies, or templates that help the AI understand the task in context.	"Tables and templates from previous projects showing risks such as delays, cost overruns, and operational challenges have been uploaded to guide risk identification."
Adjustments & Constraints	Outline any specific conditions or limitations that should be considered in the AI's response.	"Ensure compliance with security and gdpr regulations and account for technical constraints while creating the risks."
Types of Output	Define the desired format or structure for the AI's response.	"Present the risk assessment in a tabular format, with clear categorization of each risk."
Evaluation & Steps	Clarify additional checks or steps the AI should follow to ensure accuracy, completeness, and effectiveness of the response.	"Ensure all identified risks are relevant, consider regulations, and validate risk mitigation strategies to ensure they are actionable and realistic for the project."

Below is an example of a prompt using CREATE format 

Character: You’re a seasoned project manager with experience in software development.

Request: Generate a comprehensive project timeline for an upcoming software development project,

including key phases, deliverables, and deadlines.

Examples: The project involves four main phases: planning, development, testing and production deployment. Key milestones include project kickoff, completion of wireframes, alpha release, and final

product delivery.

Adjustments and constraints: Ensure that the timeline accounts for buffers for unexpected delays

Types of output: Provide the timeline in Gantt-chart format, clearly indicating the duration of each

phase and milestone

Evaluation and steps: Evaluate the timeline based on its alignment with project objectives, feasibility, and potential risks. Break down the task into steps such as gathering project requirements, drafting the

timeline, and reviewing for accuracy.

Conclusion

In the age of AI, it is very important to realize the need to adopt to the new tools being developed nearly every day, thanks to AI. One of the fundamentals is to understand how to prompt and to keep a few formulas handy and to reuse them for all your project management needs and work smarter. 

How project management can leverage AI

The Impact of AI on Project Management: Automating Routine Tasks to Boost Efficiency

Introduction

The rapid developments in the field of Artificial Intelligence (AI) are truly breathtaking. It has taken the world by storm, with incredible advancements happening at lightning speed. Even though AI-powered chatbots like ChatGPT were already around in the early 2020s, they have gained mainstream attention over the last two years. AI’s potential has expanded beyond just answering questions—it can now make phone calls, place orders, assess user needs, and much more. The possibilities seem limitless.

The same holds true for project management. AI is increasingly becoming a game-changer in this space as well. Following the rise of ChatGPT, the Project Management Institute (PMI) has developed its own AI tool specifically designed for project management, called PMI Infinity. This is an AI-powered chatbot trained with the Project Management Body of Knowledge (PMBOK) and other project management resources. PMI Infinity can assist with creating project plans, drafting project charters, and offering guidance using PMI’s vast knowledge base. I’m looking forward to trying PMI Infinity out and sharing more insights in a future post.

AI Generated Image

---

Identifying Pain Points for Project Managers

To effectively leverage AI in project management, it's crucial to identify the key pain points faced by project managers (PMs) and Project Management Offices (PMOs). These pain points often revolve around time-consuming, manual tasks that detract from focusing on higher-level strategic work. By automating repetitive tasks, AI can help PMs save time and resources, reduce human errors, and enhance the speed and accuracy of project delivery.

Here are some of the most common pain points and how AI can address them:

1. Report Generation (e.g., Weekly Status Reports, Follow-up Emails)

Pain Point: Project managers often spend a significant amount of time preparing and distributing weekly status reports, as well as sending follow-up emails to team members and stakeholders. This process can be tedious and prone to delays.

AI Solution: AI-powered tools can automate report generation, providing real-time updates based on project data. Additionally, AI can automatically send follow-up emails to keep team members informed and ensure deadlines are met without the need for manual intervention.

2. Documentation Preparation (e.g., Project Charters)

Pain Point: Creating detailed project documentation, such as project charters, can be time-consuming. It requires gathering information from various sources and ensuring that all the necessary details are included.

AI Solution: AI-driven document generation tools can create project charters in a fraction of the time it would take a PM to do manually. AI can pull from existing templates, project data, and industry best practices to streamline the process, allowing PMs to focus on more critical tasks.

3. Data Analysis and Monitoring

Pain Point: Accurate data analysis is essential for tracking project progress, budget spent, and resource utilization. However, sifting through large volumes of data to extract meaningful insights can be overwhelming.

AI Solution: AI algorithms can analyze vast amounts of project data, highlighting key trends, flagging anomalies, and predicting potential risks. Tools like AI-powered dashboards allow PMs to have real-time visibility into project health, making decision-making faster and more informed.

4. Risk Management and Risk Registry Updates

Pain Point: Managing and maintaining a comprehensive risk registry is vital to project success, but it often involves extensive manual tracking and updating. Missing a key risk or delay in updating the registry can lead to significant project setbacks.

AI Solution: AI can help project managers stay on top of risk management by automatically updating the risk registry based on project changes, identifying emerging risks, and providing recommendations on how to mitigate them. This reduces the risk of human error and keeps the project on track.

Conclusion

AI has the potential to revolutionize task management in project management. By automating routine tasks, AI enables project managers to focus on higher-value activities, such as strategic planning and stakeholder management. The pain points outlined above are just a few examples of where AI can provide immediate relief. As AI continues to evolve, project managers will have even more tools at their disposal to optimize efficiency, improve accuracy, and deliver successful projects on time.

---

Prompt Engineering and Project Management

Introduction

AI has revolutionized many industries, and project management is no exception. It's essential to understand how AI can serve as a powerful tool in our daily lives. When applied effectively, AI has the potential to be a game-changer for project managers. In the future, the workforce will be divided into those who can harness the power of AI and those who cannot. Therefore, it’s crucial to embrace AI's benefits to stay ahead.

AI won’t replace jobs but will change the type of tasks humans perform. We’ll see more human involvement in roles requiring creativity and personal interaction, leaving repetitive and automated tasks to AI. In this evolving landscape, mastering AI tools can help us focus on the more strategic and meaningful aspects of our work.

Understanding Prompts

Prompts are the instructions or questions you provide to AI chatbots, enabling them to generate responses tailored to your needs. Whether you're using tools like ChatGPT, Perplexity, or Gemini, learning the art of crafting prompts, or "prompt engineering," is vital for getting the best results.
When interacting with chatbots, one rule stands true: **Garbage In, Garbage Out.** The more refined and clear your prompts are, the more accurate and customized the AI's output will be. This makes it critical to follow certain principles when constructing your prompts.
What Is Prompt Engineering?
Prompt engineering is an essential skill that involves carefully designing, testing, and refining prompts to guide AI models toward producing accurate, relevant, and useful outcomes. Mastering this process can optimize your workflows, increase productivity, and enhance project success. Several strategies are key to effective prompt engineering:
- Crafting precise and structured prompts
- Providing relevant context
- Setting an appropriate tone
- Including examples
- Continuously experimenting and refining your prompts
- Understanding the audience
- Using clear, concise language
- Adding structure and reliability checks

In this blog, I'll break down the basics of writing prompts that will help you achieve more customized outputs. These elements are essential to ensure AI delivers responses that meet your expectations:

- Be specific about what you need
- Define the tone of the output
- Include examples for reference
- Experiment and refine with follow-up prompts
- Ensure clarity
- Provide clear context

Now, let's dive into a simple yet effective formula for crafting prompts.

The RTFS Formula

One of the most commonly used and straightforward prompt structures is the RTFS formula, which stands for Role, Task, Format, Source.
- Role: Define the role the AI should assume to guide its response.
- Task: Clearly state the task you want the AI to perform.
- Format: Specify the format you need for the output, which helps in customizing the results to your liking.
- Source: If applicable, mention specific sources the AI should consider when generating the response. This is especially crucial in enterprise settings, where internal resources may be necessary.

Example Prompt Using RTFS Formula Here’s an example of how to use the RTFS formula in practice:

- Role: You are an experienced project manager specializing in Tourism and Travel IT.
- Task: Identify potential risks in project XYZ for developing an automated check-in counter at the airport, considering factors such as delays, budget overruns, traveler feedback, and uninterrupted operations.
- Format: Provide the risks in a table format, creating a risk register.
- Source: Use "Project Solution Details" as the primary reference (this could be an internal document or webpage) and also refer to external internet-based information for similar projects.