Want to improve the content generated by AI? Wondering how top AI pros get what they need from AI?
In this article, you’ll discover a four-step AI prompt engineering framework and what to do if it doesn’t work.
What Is AI Prompt Engineering?
Prompt engineering is the art and science of communicating with large language models to elicit the desired output. It involves crafting carefully structured instructions, questions, or “prompts” that guide artificial intelligence (AI) in generating content, analyzing data, or providing recommendations that align with your specific needs and goals.
Think of prompt engineering as “programming” the generative AI using natural language processing. Just as software developers write code to instruct a computer on what complex tasks to perform, AI prompt engineers use words and sentences to direct the AI model toward producing the content or insights they want.
The key to being a savvy AI prompt engineer lies in providing the AI with clear, specific, and well-organized information that helps it understand the relevant context, purpose, and desired outcome of the task at hand. The more relevant and detailed your prompts, the better the AI can comprehend and fulfill your request.
#1: Apply the ‘RACE’ Framework for AI Prompt Engineering
To help you with your prompt engineering techniques, Chris Penn, co-founder of Trust Insights and a leading expert in AI for marketing, recommends using the “RACE” framework:
R – Role
The first step in crafting an effective prompt is to define the role you want the AI to assume. This helps the model understand the perspective, tone, and level of expertise it should adopt when generating content.
For example, if you're looking to create a blog post about the benefits of your product, you might start your prompt with:
- “Act as an experienced content marketer specializing in writing persuasive blog posts highlighting software products' unique value propositions.”
By setting the role upfront, you're priming the AI to generate content that aligns with the desired perspective and expertise.
A – Action
Next, specify the action or task you want the AI to perform. This is where you provide explicit instructions on what type of content you need the AI to generate.
Continuing with our blog post example, you might say:
- “Your task is to write a compelling, 1,000-word blog post that showcases the top 5 benefits of our new project management software for small businesses.”
The more specific and detailed your action instructions are, the better the AI can understand and deliver the content you're looking for.
C – Context
Context is king when it comes to prompt engineering. The more background information and relevant details you can provide the AI about your brand, audience, industry, and specific goals, the more tailored and effective the generated content will be.
For our blog post prompt, you might include additional context like:
Ready to Supercharge Your Marketing Strategy?
Get expert training and an unbeatable conference experience when you attend Social Media Marketing World—from your friends at Social Media Examiner.
Broaden your reach, skyrocket your engagement, and grow your sales. Become the marketing hero your company or clients need!
🔥 As a valued reader, you can save $700 on an All-Access ticket if you act now. Sale Ends Friday! 🔥
GET THE DETAILS- “Our target audience is small business owners and entrepreneurs looking to streamline operations and boost productivity.
- Our brand voice is professional, informative, and empathetic. We aim to educate and inspire our readers while positioning our product as a valuable solution to their challenges.
- The blog post should focus on our software's five key benefits: improved team collaboration, enhanced project visibility, streamlined task management, automated reporting, and seamless integrations with other tools.
- Please include real-world examples, statistics, and customer testimonials to support the benefits mentioned.”
By providing this rich context, you're giving the AI a wealth of information, enabling it to generate a blog post that truly speaks to your target audience and aligns with your brand messaging.
E – Execute
The final step in the “RACE” framework is to specify the desired format and any additional execution instructions for the AI-generated content.
For our blog post example, you might say:
- “Please write the blog post in a conversational, engaging style that's easy to read and understand.
- Structure the post with an introduction, five main sections (one for each benefit), and a conclusion that summarizes the key points and includes a strong call to action.
- Use headings, subheadings, bullet points, and short paragraphs to break up the text and improve readability.
- Include relevant images, charts, or infographics to support the content visually (you can use placeholders for now).
- Optimize the post for SEO by naturally incorporating relevant keywords and phrases and including meta tags and descriptions.”
By carefully designing your prompts using the “RACE” framework, you'll be able to collaborate with AI more effectively and generate content that truly resonates with your audience and drives your marketing goals.
#2: Optimize AI-Generated Content With the ‘PARE’ Framework for AI Prompt Engineering
While the “RACE” framework provides a solid foundation for prompt engineering, there may be times when you need to go beyond the initial prompt to refine and optimize the AI-generated content. This is where the “PARE” framework comes into play, offering a structured approach to iterating and improving the AI's output until it meets your high standards.
P – Prime
The first step in the “PARE” framework is to prime the AI with additional context and information relevant to the task at hand. This step involves asking the AI to share its existing knowledge and understanding of the topic, which can help surface key concepts, ideas, and themes to incorporate into the content.
For example, if you're working on a social media campaign for a new sustainable fashion brand, you might prime the AI by asking:
- “What do you know about the current trends and challenges in the sustainable fashion industry? Please share some key insights and statistics that could be relevant to our campaign.”
The AI's response might include information about the growing consumer demand for eco-friendly clothing, the environmental impact of fast fashion, and examples of successful sustainable fashion brands and their marketing strategies.
By priming the AI in this way, you're helping to activate its knowledge base and bring meaningful context to the forefront, which can inform and enhance the content it generates.
A – Augment
Once you've primed the AI, the next step is to augment its understanding by providing additional information, clarifications, or requirements based on its initial response.
In our sustainable fashion example, you might review the AI's priming response and identify areas where you can provide more specific guidance, such as:
“Thanks for those insights! To make our campaign even more targeted, please focus on the following key aspects:
Curious About How to Use AI?
Our newest show, AI Explored, might be just what you're looking for. It's for marketers, creators, and entrepreneurs who want to understand how to use AI in their business.
It's hosted by Michael Stelzner and explores this exciting new frontier in easy-to-understand terms.
Pull up your favorite podcast app and search for AI Explored. Or click the button below for more information.
- The importance of transparency and traceability in sustainable fashion supply chains
- The role of influencer marketing in promoting sustainable fashion to millennials and Gen Z
- Specific examples of how our brand minimizes its environmental footprint through innovative materials and production processes.”
By augmenting the AI's understanding with this additional context, you're helping to refine its knowledge base and steer it towards generating content that's even more relevant and aligned with your campaign goals.
R – Refresh
After you've primed and augmented the AI, it's time to put it to work, generating the actual content for your campaign. However, it's unlikely that the AI will nail it on the first attempt. That's where the refresh stage comes in.
Refresh involves reviewing the AI-generated content, identifying areas for improvement, and providing feedback to the AI on how to refine its output.
For our sustainable fashion campaign, you might receive an initial batch of social media posts from the AI and notice that while the content is informative, it lacks the engaging storytelling element crucial for capturing attention on social media.
To refresh the AI's approach, you could provide the following feedback:
“Thanks for the initial posts! They're packed with great information, but we need to make them more engaging and emotionally compelling to stop scrollers in their tracks. Please try weaving in more personal stories, aspirational messaging, and powerful calls to action that tap into our audience's values and desires. For example, you could highlight the story of one of our artisans and how their life has been positively impacted by working with our brand, or paint a picture of the sustainable fashion future we're working towards and invite our followers to be part of the movement.”
By providing this kind of specific, actionable feedback, you're guiding the AI to iterate and improve upon its initial output, bringing it closer to the high-impact content you're looking for.
E – Evaluate
The final step in the “PARE” framework is to evaluate the refined content against your original goals and criteria. This involves closely reviewing the AI's latest output to ensure it meets your standards for quality, relevance, and alignment with your brand voice and messaging.
In our sustainable fashion example, you would carefully read through the refreshed social media posts and ask yourself questions like:
- Does this content effectively communicate our brand's unique value proposition and commitment to sustainability?
- Is the messaging compelling, emotionally resonant, and likely to inspire engagement and action from our target audience?
- Are the posts visually appealing, with strong imagery and formatting that aligns with our brand aesthetic?
- Is the content optimized for each social media platform, with appropriate hashtags, tags, and calls to action?
If the answer to all these questions is a resounding “yes,” congratulations—you've successfully collaborated with the AI to create high-impact content for your campaign! If there are still areas for improvement, repeat the refresh and evaluate stages until you're fully satisfied with the results.
By following the “PARE” framework, you can take your prompt engineering skills to the next level, iterating and refining your AI-generated content until it truly shines. This approach ensures that you're not just settling for the AI's first attempt but working closely with the technology to create content that exceeds your expectations and drives real results for your brand.
#3: 7 More Tips for Effective Prompt Engineering
Now that you've got a solid grasp on the “RACE” and “PARE” frameworks, let's dive into some additional tips and best practices to help you optimize prompts:
Be as Specific and Detailed as Possible: The more specific and detailed your prompts are, the better the AI will understand and deliver the content you're looking for. Don't hesitate to provide lengthy, in-depth context and instructions—the AI can handle it!
Use Examples to Clarify Your Expectations: One of the best ways to help the AI understand what you're looking for is to provide examples of the kind of content you want it to generate. This could be links to blog posts, social media updates, or any other relevant content that embodies the style, tone, and messaging you're aiming for.
Experiment With Different Prompts and Approaches: Prompt engineering is as much an art as a science, and what works for one brand or campaign may not work for another. Don't be afraid to experiment with different prompts, frameworks, and approaches to see what yields the best results for your specific needs.
Collaborate With Your Team: Prompt engineering is a collaborative process, and the more minds you have working together to craft effective prompts, the better the results will be. Encourage your team members to contribute their ideas, insights, and feedback throughout the prompt engineering process.
Keep the End Goal in Mind: Always keep your ultimate marketing goals and objectives front and center when crafting your prompts. Every piece of context, instruction, or feedback you provide should serve to create content that drives your desired outcomes, whether that's increased brand awareness, engagement, conversions, or customer loyalty.
Use AI-Generated Content as a Starting Point: While AI models can generate some truly impressive content, it's important to remember that it's not a complete replacement for human creativity and judgment. Use AI-generated content as a starting point, but always take the time to review, edit, and refine it to ensure it truly meets your standards and aligns with your brand voice.
Stay Up-to-Date on the Latest Developments: The field of AI and prompt engineering is constantly evolving, with new models, techniques, and best practices emerging continually. To stay ahead of the curve, regularly read up on the latest developments, attend industry conferences and webinars, and connect with other prompt engineering professionals to share knowledge and ideas.
Chris Penn is a data scientist and author of AI for Marketers: An Introduction and Primer. He is also co-founder and Chief Data Scientist of Trust Insights, a consultancy that helps brands with analytics and AI. His course is Generative AI for Marketers. You can find him on LinkedIn and Threads.
Other Notes From This Episode
- Connect with Michael Stelzner @Stelzner on Facebook and @Mike_Stelzner on X.
- Watch this interview and other exclusive content from Social Media Examiner on YouTube.
Listen to the Podcast Now
This article is sourced from the AI Explored podcast. Listen or subscribe below.
Where to subscribe: Apple Podcasts | Spotify | YouTube Music | YouTube | Amazon Music | RSS
✋🏽 If you enjoyed this episode of the AI Explored podcast, please head over to Apple Podcasts, leave a rating, write a review, and subscribe.
Stay Up-to-Date: Get New Marketing Articles Delivered to You!
Don't miss out on upcoming social media marketing insights and strategies! Sign up to receive notifications when we publish new articles on Social Media Examiner. Our expertly crafted content will help you stay ahead of the curve and drive results for your business. Click the link below to sign up now and receive our annual report!
Discover Proven Marketing Strategies and Tips
Want to go even deeper with your marketing? Check out the Social Media Marketing Podcast! Publishing weekly since 2012, the Social Media Marketing Podcast helps you navigate the constantly changing marketing jungle, with expert interviews from marketing pros.
But don’t let the name fool you. This show is about a lot more than just social media marketing. With over 600 episodes and millions of downloads each year, this show has been a trusted source for marketers for well over a decade.