Use AI like an expert
This week's task shows you how to prompt engineer with reusable templates for you and your team
Generative AI has become part of knowledge workers daily tasks, both for their companies and in their personal lives.
But many of us are not making the most of them.
Aside from what tool you are using and whether it has access to the right data, the biggest improvement you can make is the instructions you provide.
This is called prompt engineering, a way to tell AI tools (chatbots or writing assistants) exactly how to help with tasks, like answering customer questions, writing reports, or even brainstorming ideas.
Doing it properly means your results are fast, useful, and save you time.
By the end of today’s task you will be able to give your AI tools instructions that will give you and your team the best answers.
Prompt engineering for knowledge workers
Today’s task provides you with the essential skills in prompt engineering to help you effectively communicate with AI tools.
Through three practical steps, you'll learn how to craft prompts that produce useful and accurate responses from AI systems like ChatGPT.
Each section includes templates and examples designed to build your confidence.
Introduction to prompt engineering
Think of prompt engineering as learning how to communicate effectively with artificial intelligence.
Just as clear communication is important when working with colleagues, well-designed prompts help you get better results from AI.
This skill has become increasingly valuable for knowledge workers across all industries who use AI systems to enhance productivity.
When interacting with AI models, the quality of your input directly affects the quality of the output you receive.
Prompt engineering helps bridge the gap between what you want and what the AI delivers.
The good news is that prompt engineering doesn't require technical expertise or programming knowledge. Instead, it's about understanding how to structure your requests in ways that AI systems can best interpret and respond to.
Todays task is designed for all knowledge workers, regardless of previous experience with AI.
We will start with basic concepts and progressively build toward more advanced techniques.
By the end, you'll have practical templates you can apply immediately to improve your work processes and outcomes when using AI assistants.
Step 1: Basic prompt construction
Learning objectives
In this first step, you will learn the fundamental components of effective prompts.
You will understand how to structure clear instructions that generate useful responses from AI systems.
By the end of this section, you will be able to create basic prompts that communicate your needs effectively to AI tools.
Key components of a prompt
A well-crafted prompt typically contains several essential elements that help the AI understand what you're asking for.
The primary component is the directive, which is a clear instruction that tells the AI exactly what task it should perform.
Without a clear directive, the AI may provide generic or irrelevant responses.
Directives work best when they use action verbs and specific language that leaves little room for misinterpretation.
Another important component is a role assignment.
This tells the AI what perspective or expertise to adopt when responding.
For example, you might ask it to "respond as a marketing specialist" or "explain this like a teacher would to a beginner."
The role helps frame how the AI approaches your request and can significantly influence the style and content of the response.
Context information provides the AI with background details necessary to generate an appropriate response.
This might include relevant facts, constraints, or specifications that help the AI understand the full picture of what you're asking.
Including the right context can make the difference between a generic answer and one that is precisely tailored to your needs.
Step 1 instructions
Start by identifying exactly what information or assistance you need from the AI.
Formulate a clear directive using specific action verbs (explain, summarise, list, analyse).
Consider whether assigning a role would help frame the response appropriately.
Add any necessary context information the AI would need to produce a relevant response.
Review your prompt for clarity and specificity before submitting it.
Basic prompt template
I need [specific information or assistance].
Please [action verb] [specific subject or topic].
[Include any relevant context or constraints].
[Optional: Present the information in a specific format].Example
I need to understand the concept of supply chain management.
Please explain the basic principles of supply chain management.
I have limited knowledge of logistics, so use simple terms and everyday examples.
Present the information in short paragraphs with bullet points for key concepts.Practice exercise
Try creating three different prompts using the template above for tasks relevant to your work.
Experiment with different directives and see how changing the wording affects the responses you receive.
Any questions, contact me via direct message or as a subscriber add your comments below.
Step 2: Advanced prompt techniques
Learning objectives
In this second step, you will learn more sophisticated techniques to enhance your prompts.
You will discover how to use examples, structure output formats, and incorporate specificity to get precisely tailored responses.
By the end of this section, you will be able to craft prompts that generate higher-quality, more relevant outputs.
Advanced techniques
One powerful technique is to provide examples in your prompts, known as "few-shot" or "one-shot" prompting.
By showing the AI a sample of the type of response you are looking for, you help it understand your expectations more precisely.
Examples are particularly useful when you need the AI to follow a specific format or approach that might be difficult to explain in words alone.
Another advanced technique is to specify the desired output format.
By clearly indicating how you want information structured, whether as a list, table, step-by-step guide, or narrative, you make the response more useful for your specific needs.
Being explicit about format saves you time reformatting later and helps the AI organise information in the most accessible way.
Adding constraints or parameters helps refine the AI's response further.
You might specify word count limits, reading level, tone, or perspective. These constraints act like guardrails that keep the AI focused on delivering exactly what you need rather than generic information that may require further editing.
Step 2 instructions
Identify your specific information need and desired outcome.
Determine if providing an example would help clarify your instructions.
Specify any formatting requirements for the output.
Add relevant constraints regarding length, style, or approach.
Include any special instructions about how to handle the information.
Advanced prompt template
I need [specific information or assistance].
Please [action verb] [specific subject or topic].
Here's an example of what I'm looking for: [Provide a sample of the desired output format or approach]
Format the response as [desired structure: bullet points, table, paragraphs, etc.].
Additional requirements:
- [Constraint 1, e.g., "Keep explanations under 100 words per point"]
- [Constraint 2, e.g., "Use non-technical language"]
- [Constraint 3, e.g., "Include real-world applications"]Example
I need to create a comparative analysis of remote work policies.
Please analyse the pros and cons of different remote work approaches for a mid-sized technology company based in London.
Here is an example of what I'm looking for:
Fully Remote Model:
Pros: Wider talent pool, reduced office costs, increased employee satisfaction
Cons: Potential communication challenges, difficulty building company culture
Format the response as a comparison table with three models: fully remote, hybrid, and office-centric.
Additional requirements:
- Include implementation considerations for each model
- Focus on impacts on productivity, employee satisfaction, and operational costs
- Provide 1-2 examples of companies successfully using each modelPractice exercise
Create a prompt using the advanced template for a complex task in your field.
Include an example of the output format you want and at least three specific constraints or requirements.
Test how the AI responds and refine your prompt if needed.
Any questions, contact me via direct message or as a subscriber add your comments below.
Step 3: Creating reusable prompt templates
Learning objectives
In this final step, you will learn how to develop reusable prompt templates for recurring work scenarios.
You will understand how to analyse tasks and create flexible templates that can be adapted for different situations.
By the end of this section, you'll be able to build a personal library of prompt templates that increase your efficiency.
Developing reusable templates
Reusable prompt templates are pre-defined structures that can be filled with specific content for efficient generation of new prompts.
They are particularly valuable for knowledge workers who frequently perform similar types of tasks with AI assistance.
Creating your own library of templates can dramatically reduce the time spent crafting prompts and ensure consistency in the outputs you receive.
This library can be shared with colleagues to create more efficient ways of working for your team.
The process begins with analysing the common tasks you perform and identifying patterns in the types of information you typically request.
Look for tasks that follow similar workflows or require similar types of outputs.
By recognising these patterns, you can design template structures that accommodate various inputs while maintaining a consistent framework for the AI to follow.
Testing and refining your templates is an essential step in the development process.
A template that works perfectly for one scenario might need adjustments for another.
Plan to iterate on your templates based on the results you receive, gradually improving their versatility and effectiveness.
The best templates evolve through practical application and thoughtful revision.
Step 3 instructions
Identify recurring tasks or information needs in your work.
Analyse the common elements across different instances of these tasks.
Create a template structure with placeholders for variable information.
Test the template with different inputs to ensure flexibility.
Refine the template based on results and save it for future use.
Template development framework
Task Purpose:
[Describe the general purpose this template will serve]
Template Structure:
[Directive with placeholders for variables]
[Context section with placeholders]
[Output format specifications]
[Standard constraints or requirements]
Variables to Replace:
- [Variable 1]: [Description of what should replace this variable]
- [Variable 2]: [Description of what should replace this variable]
- [Variable 3]: [Description of what should replace this variable]
Example Implementation:
[Show a completed example of the template in use]Example
Task Purpose:
Creating concise summaries of research articles for team knowledge sharing
Template Structure:
I need a summary of the research article titled "[ARTICLE TITLE]".
Please analyse this [FIELD] article and provide a concise summary highlighting the key findings, methodology, and practical implications.
Format the response as follows:
- Research Question/Objective (1-2 sentences)
- Methodology (2-3 bullet points)
- Key Findings (3-5 bullet points)
- Practical Implications (2-3 sentences)
- Limitations (1-2 sentences)
Additional requirements:
- Use language appropriate for [AUDIENCE LEVEL]
- Keep the entire summary under [WORD COUNT] words
- Highlight any connections to [RELATED TOPIC/PROJECT]
Variables to Replace:
- [ARTICLE TITLE]: Insert the full title of the research article
- [FIELD]: Specify the field of research (e.g., marketing, psychology, data science)
- [AUDIENCE LEVEL]: Specify expertise level (e.g., beginner, intermediate, expert)
- [WORD COUNT]: Specify desired length (typically 300-500 words)
- [RELATED TOPIC/PROJECT]: Mention specific projects or topics relevant to your team
Example Implementation:
I need a summary of the research article titled "Impact of Social Media Marketing on Consumer Behaviour in Emerging Markets".
Please analyse this marketing article and provide a concise summary highlighting the key findings, methodology, and practical implications.
Format the response as follows:
- Research Question/Objective (1-2 sentences)
- Methodology (2-3 bullet points)
- Key Findings (3-5 bullet points)
- Practical Implications (2-3 sentences)
- Limitations (1-2 sentences)
Additional requirements:
- Use language appropriate for beginner marketing professionals
- Keep the entire summary under 400 words
- Highlight any connections to our upcoming social media campaign in Southeast AsiaPractice exercise
Identify a recurring task in your work that could benefit from AI assistance.
Use the template development framework to create a reusable prompt template for this step.
Test your template with at least two different sets of variables and refine it based on the results.
Any questions, contact me via direct message or as a subscriber add your comments below.
Conclusion
Throughout this task, you have gained valuable skills in prompt engineering that can transform how you and your team work with AI tools.
Beginning with basic prompt construction, you learned the essential components that make up effective prompts, including clear directives and appropriate context.
You then advanced to more sophisticated techniques like using examples and specifying output formats to get precisely tailored responses.
Finally, you developed the ability to create your own reusable templates for recurring tasks.
The templates and frameworks provided in each task serve as practical guides that you can adapt to your specific work scenarios.
By applying these tools consistently, you will notice significant improvements in the quality and relevance of AI-generated outputs.
Remember that prompt engineering is both an art and a science, it benefits from experimentation and refinement over time.
The more you practice, the more intuitive the process will become.
As AI continues to evolve as a productivity tool for knowledge workers, your ability to craft effective prompts will become an increasingly valuable skill.
The techniques you've learned here will help you communicate more efficiently with AI systems, saving time and improving outcomes across various aspects of your work.
Keep experimenting with different approaches, and don't hesitate to refine your prompts based on the results you receive.
If you feel you can benefit from these AI Tasks, with the other subscriber benefits of weekly AI Currents, the weekly industry round-up of news, product launches and M&A, then please click to subscribe.