AI Prompts for Content Marketing

AI is changing everything about how we do business, and content marketing is no different. Many content marketers have experimented with tools like Chat GPT or Bard, but how do you incorporate this technology into your workflow? AI can be used in every aspect of content marketing strategy, planning and production — and in the near future, it will be. If you’re ready to get a headstart, this guide is for you.

In this guide, we cover: 

• Key Terms to Know 
• How to Write a Good Prompt
• Prompt Engineering Best Practices 
• AI Prompts for Audience Research
• AI Prompts for Content Goals & Objectives 
• AI Prompts for Content Audit & Analysis 
• AI Prompts for Content Strategy & Planning 
• AI Prompts for Content Production
• AI Prompts for Distribution & Promotion 
• AI Prompts for Measurement & Optimization

Ready? Let’s dive in.

Disclaimer

• Please review all AI-generated content, as all AI tools occasionally output incorrect information. 
• Never share private or sensitive data or confidential information with a Large Language Model. 

AI & Content Marketing: What to Know 

We’re going to skip a full explanation of what artificial intelligence is — there are plenty of resources covering this topic. A great place to start is Intro to AI for Marketers, a free class offered regularly by the Marketing AI Institute. 

Key Terms to Know 

A few terms we reference in this guide are: 

• AI: Artificial intelligence is an umbrella term used for the technologies that aim to give machines human-like intelligence, based on our understanding of how the human mind works. The most prominent form of AI today is machine learning, which is the ability for the AI to learn based on past interactions and create its own path to achieve a desired outcome.  
• Large Language Model (LLM): A deep learning model (a subset of AI and machine learning) that understands and generates language. Some of the most common LLMs right now are ChatGPT, Bard and Claude, and use a chat interface.
• Prompt: With Large Language Models, a prompt is the input (or series of inputs) a user enters that triggers a response from the AI. This might be a simple phrase or sentence (i.e. “explain particle physics at a third-grade level”), or thousands of words of context.
• Prompt Engineering: Writing prompts for LLMs and testing and refining these prompts. Users may need to adjust prompts to get better results, and they should revise prompts over time as AI tools become more sophisticated. 
• Prompt Library: Any repository of prompts used with AI, so tried and tested prompts can be reused by you or another person on your team.  

Prompts are a critical piece of working effectively with AI today. With LLMs, the prompt is how you communicate with the AI. This is also why AI tools like ChatGPT are so disruptive — it now gives anyone the ability to interact with and instruct a LLM through natural language.

So let’s discuss what a good prompt looks like.  

Prompt Engineering: How to Write a Good Prompt 

As anyone who’s experimented with a chat-based AI knows, not all prompts are created equal. Short, generic prompts tend to yield lackluster, generic outputs. 

Does that mean that AI is ineffective? Not necessarily. A better version of the same prompt will result in a wildly different outcome. Let's look at an example.

A Bad Example of Prompt Engineering

A Good Example of Prompt Engineering

Now, the rules of what makes a good prompt will probably continue to evolve as AI becomes more sophisticated and as it learns from the training data of past users — the more information it’s fed, the more data the AI has to work with. (We're also assuming you're working with an "out of the box" LLM, and not one that's been fine-tuned to your specific use case.)

All that to say, the AI is reliant on us to offer enough context about the task we’re asking it to do. Consider: If you asked your intern to “write a blog post,” they might ask for more information to complete the task. Things like: 

• Who’s the target audience? 
• What’s the goal? 
• What length should the blog post be? 
• What tone do you prefer? Etc.

As Christopher Penn of Trust Insights explains, effective LLM prompts include four components: Role, Action, Context and Execute (“R.A.C.E.”). 

A "R.A.C.E." prompt produces a better output because it more clearly outlines the expectations you have for that output. Let's look at another example.

In the example above, the prompt begins by defining the AI's role: as a content marketing strategist. There's been some debate about this strategy — whether it's really relevant and necessary to give the AI a job like "you are an accomplished copywriter" or "you are a rockstar sales executive."

But this tactic isn't about inflating the AI's ego; it's about semantic relevance. LLMs are predictors; they predict words based on a highly sophisticated understanding of how words relate to one another. (This is part of topic modeling.)

So by defining the AI's role, we are offering additional context that "expert content marketing strategist" is relevant to this task.

The second portion of this prompt outlines the Action; this is where the goal of the prompt is really defined. Experiment with giving more and less direction here; the AI's output might include ideas or a format you hadn't previously considered. This is part of the benefit of leveraging machine learning: an ability to go "above and beyond" what traditional software would produce.

But include any important details or guidelines that you want the AI to follow. In this example, the prompt asks the content calendar to be structured into a table.

Next is Context, which I'd argue is the most critical. The more information you can give the AI, the more specific and relevant its output will be. This is where sophisticated prompt engineering comes in, and your opportunity to inject your own intelligence, creativity, expertise and approach into the process.

The context section could be even longer than what's outlined in this example — depending on the prompt word count limit (which is related to the token limits for that model), you could include hundreds or even thousands of words of context. For example, instead of the few sentence summary of this company's current content strategy, you could include the company's entire blog in the prompt.

Lastly, we have Execute, or instructing the AI to complete the prompt as outlined.

Good AI Prompts 

Keep the following in mind whenever you prompt a chat-based AI: 

1. Give the AI context. The more context you give it, the better your response will be. Be descriptive and use a variety of words and synonyms. 
2. Ask if it needs more information, or if it has any questions for you before it begins. This can help identify any gaps in your prompt. 
3. Provide feedback. Professor Ethan Mollick recommends that users resist the urge to limit your exchange to a single, all-encompassing prompt — and expecting the AI to complete the task on the first try. Instead, treat the exchange like a conversation, providing feedback and redirecting to get a more appropriate response. 

Prompt Engineering Best Practices 

No matter what your goal is, keep the following prompt engineering best practices in mind:

• Never include sensitive information in a prompt.
• Always review the output of AI. AI occasionally output inaccurate information, so all results should be double-checked by a subject matter expert. 
• Save prompts that work well for you in a prompt library. This can be anywhere you store knowledge: A Google Doc or Sheet, Evernote, Notion, etc. 

In the following sections, we'll detail many of the ways you can use AI in your content marketing program — from research and planning to production and reporting. Later on, we'll cover why to use AI, when to use it, and when not to use it.

Audience Research

The first step in any content marketing strategy is to identify your brand's target audience. This is central to understanding the buyer journey and common questions buyers have — which dictate the content that will be most helpful to prospective customers.

Key components of understanding your audience include: 

• Buyer Personas
• Customer Journey Mapping
• Needs & Pain Points

Buyer Personas

AI can be used to: 

• Research information about particular cohorts or buying groups to inform buyer personas. 
• Generate rich buyer persona profiles based on information. 
• Role-play as a buyer persona. Share information about the buyer persona, then ask questions about its motivations, preferences or feedback on marketing campaigns.  
• Outline common questions this buyer persona might ask.

Keep in mind that AI doesn't know your customers — it doesn't know anything. It is merely predicting words based on the information its been fed (which may include legitimate research, anecdotes, feedback, etc. from your audience or the cohort you're targeting).

AI may not entirely replace customer interviews and buyer persona research, but it can certainly identify patterns we may not see, uncover information we might have missed, raise questions we might benefit from asking, and develop a richer persona we might use to empathize with our target customers.

In the past, a marketing agency or content marketing team may have created fictional buyer personas that include writing in the buyer's "own words." This would be based on research, but might be biased based on the team's assumptions about these consumers. AI can help give us alternative perspectives, uncover blindspots in our strategic thinking, and give us new language, visuals and ideas to connect with and convert these customers.

Needs & Pain Points

Understanding customer needs and pain points is one of the most important aspects of any business, and it’s essential to writing relevant, empathic content. 

AI can be used to: 

• Brainstorm needs and pain points based on buyer persona information. 
• Summarize buyer needs and pain points based on existing information — for example, customer reviews, contact form submissions or customer service transcripts.  
• Perform sentiment analysis on existing blog post comments, to better understand whether existing content meets buyer needs and answers their questions.
• List common questions found in customer or search query data to identify new content ideas.
• Analyze how well your existing content addresses your buyer personas' needs and motivations. (Orbit Media offers a great example with their competitive matrix heat map.)

In the example prompt above, we ask AI to outline the primary needs, pain points and motivations of a buyer persona. Keep in mind that when using AI for buyer research, there are a few approaches content marketers can take:

• Freewrite: Outside of a general description or definition of the target audience, this prompt would be the most broad. The marketer gives AI the freedom to respond without restrictions. This approach would provide the most creativity and potentially uncover blindspots that marketers may have about their customer. However, the responses would need to be verified for accuracy and checked for bias.
• Analysis: This approach would be the most restrictive, and the prompt would instruct AI to identify motivations or characteristics based on existing information fed to the LLM in advance — outside research, survey results, customer interviews, chat transcripts, etc. This approach would most accurately reflect your audience, but would be limited to your existing audience and data, and marketers must be wary of protecting any sensitive or confidential information.
• Hybrid: The example above is a hybrid approach. The LLM already knows a little about the buyer persona based on a couple-paragraph description that was supplied beforehand. But the prompt includes "Rely on what I've told you and use what you know about this industry and role to create your responses (including statistics, studies, etc.)." This broadens the output to any other information about this persona that might exist in the LLM's training data. Similarly, the addition of "Otherwise, be creative and make any assumptions you need to" gives AI some freedom to expand on what's been supplied.

The approach you choose will depend on your goals, available data, and risk tolerance for sharing owned data with AI.

Get our prompts for audience research: AI prompts for content marketing professionals.

Content Goals & Objectives

Are you focused on increasing search traffic? Growing your subscriber list? Capturing qualified leads? You have to define your goals and objectives to have an effective content marketing strategy.

This includes: 

• Goals
• KPIs

Goals

AI can be used to: 

• Help define goals and objectives based on your content marketing vision and values.
• Brainstorm possible content marketing goals based on your business’s goals. 
• Suggest ideas to make your content marketing goal SMART (Specific, Measurable, Achievable, Relevant, Time-Bound). 
• Draft emails or talking points to articulate content marketing goals to other business leaders.

KPIs

AI can help:

• Classify your marketing KPIs based on goal (for example, Top-Funnel KPIs, Mid-Funnel KPIs and Bottom-Funnel KPIs). 
• List KPIs that align with your content marketing or business goals.
• Research industry standards and competitor performance to establish realistic benchmarks for KPIs.
• Design a plan to track and monitor goals and KPIs, including which tools and methods to use.

Get our prompts for content goals and objectives: AI prompts for content marketing professionals.

Content Audit & Analysis

No matter what content you have in place, it's an asset. It can provide ideas for topics, categories and themes; help you understand what resonates and what doesn't; and give you a valuable starting point for repurposing and remixing content.

This includes: 

• Existing Content Audit 
• Performance Analysis
• Competitor Analysis

Existing Content Audit 

AI can be used to:

• Identify content themes in your existing content.  
• Describe your brand tone of voice.
• Build a library of the most commonly used terms in your existing content. 

Performance Analysis

AI can be used to: 

• Identify elements of high-performing content, such as the length, style, format or topic. 
• Create dashboards, tables and charts to visualize content performance and your team's progress towards goals and KPIs.
• Analyze performance data and articulate insights, trends and key takeaways.
• Suggest copy and visuals to make performance data more understandable and engaging to key stakeholders.
• Recommend adjustments to strategy based on analytics data.

Competitor Analysis

AI can be used to: 

• Identify competitors in your market based on various criteria: search volume, market share, audience size, etc.
• Isolate elements of a competitor's high-performing content, such as the length, style, format or topic. 
• Compare and contrast competing content based on various characteristics: SEO, persuasiveness, comprehensiveness, etc.

Get our prompts for content audits and analysis: AI prompts for content marketing professionals.

Content Strategy & Planning

Armed with a clear vision for your content marketing program, you're ready to create your company's content strategy. AI can help you:

• Draft a content strategy based on what you've learned through content audit, buyer persona creation, and goal-setting exercises.
• Evaluate an existing content strategy for weaknesses or gaps.
• Create a compelling narrative to pitch your content strategy to business leaders and executives.

This also includes:

• Content Calendar 
• Content Categories & Topic Ideation

Content Calendar 

AI can be used to: 

• Organize content topics into a calendar based on your desired content categories, publishing frequency and editorial timelines. 
• Identify key seasonal events, holidays, or industry-related milestones that could impact your content calendar.
• Suggest ways to tailor or repurpose content topics to suit different customer personas or market segments, and ensure it resonates with the right target group.
• Schedule communication and production timelines when collaborating with influencers, guest authors, or other company departments.

Content Categories & Topic Ideation

AI can be used to: 

• Categorize a list of topics into content categories. 
• Brainstorm topic ideas based on a list of content categories. 
• Suggest content types (i.e. listicle, roundup, long-form, Q&A, etc.) best suited to different topics. 
• Identify thought leadership topics and themes for company leaders, based on their bios.
• Brainstorm topic and pitch ideas for thought leaders based on their area of expertise and the target outlet's audience.

Get our prompts for content strategy and planning: AI prompts for content marketing professionals.

Content Production

It's time to start churning out content. This is probably the most obvious and common use of AI today: as a writing replacement. That said, AI can play a much larger — and arguably, more impactful — role in a marketing team's content production.

This includes: 

• Editorial Process
• Writing & Editing

Editorial Process

AI can be used to: 

• Outline a suggested editorial process based on internal roles (i.e. freelance writer, content marketing manager, CMO).
• Suggest solutions to common roadblocks, bottlenecks or other issues your organization faces when sticking to content production timelines.
• Build timelines for content production based on target publishing dates and estimated turnaround times. 
• Expand editorial team roles into specific task lists. Operationalize your editorial process by outlining systems and processes.

Writing & Editing

AI can be used to: 

• Suggest prompts for content creation. 
• Organize notes into an outline. 
• Write content at scale.
• Rewrite content in different tones, lengths, styles (i.e. persuasive, educational or empathetic), and more. 
• Repurpose content into different formats:  social media posts, video and podcast scripts, newsletter summaries, etc.
• Optimize website content for search.
• Transcribe podcasts or videos to republish full transcripts, excerpts, or key points and takeaways.
• Add HTML tags to content to make formatting, uploading and publishing more efficient.

Remember: Any content generated by AI cannot be legally protected by U.S. copyright. Conduct your own independent research on the copyright and privacy laws in your region to ensure compliance and decide which content marketing assets you wish to be AI-generated. 

Get our prompts for content production: AI prompts for content marketing professionals.

Measurement & Optimization

Evaluate, optimize and pivot your content strategy through sound analytics and measurement practices.

This includes: 

• Analytics & Reporting
• Performance Management

Performance Management

AI can be used to: 

• Suggest ways to pivot your content strategy if KPIs are tracking below you goal.
• Suggest a process to regularly gather feedback from stakeholders and/or your audience to understand. Find out what is resonating and what needs improvement through surveys, comments, or check-in meetings. Then use AI to summarize insights can be integrated into content strategy.

Get our prompts for measurement and optimization: AI prompts for content marketing professionals.

Compliance & Best Practices

Industries like healthcare, financial services and telecommunications must adhere to legal and ethical guidelines. This can be challenging for content and communications teams to navigate, especially across larger organizations and global companies.

This includes: 

• Legal Compliance & Ethical Guidelines

Legal Compliance & Ethical Guidelines

AI can be used to: 

• Devise a systematic approach to review content for legal and compliance issues — aligning with legal teams, regulatory bodies, or ensuring adherence to specific industry standards.
• Outline a list of examples, dos and don'ts of your company's guidelines and policies, to help team members better understand and adhere to them.

Get our prompts for compliance and best practices: AI prompts for content marketing professionals.

‍

About Randall Pine 

Randall Pine is a marketing consulting and services agency that grows small- and mid-sized B2B companies. Since 2017, we've helped marketing teams and business leaders use content, data and marketing technology to find more (and better) leads. 

Our agency specializes in digital marketing strategy, inbound marketing, marketing automation, content, SEO and lead generation. Whether you need an ongoing partnership or to outsource a project, we can help.

‍