A report from TechCabal indicates that approximately a third of newsrooms feel prepared for the challenges of AI adoption. The same report mentions that over 75% of media companies surveyed use AI in news gathering, production, or distribution, and about a third have or are developing an institutional AI strategy. 

Mastering the art of prompt engineering is becoming a critical skill for professionals in the industry. Effective prompts can help you generate engaging headlines, outline blog posts, maintain consistency in voice and style, and even plan your content calendar.

In this article, we'll explore the 10 prompts every publishing professional should know to harness the power of AI and stay ahead of the curve. 

What Is Prompt Engineering?

Just as better ingredients can make for a better dinner, better inputs into a Generative AI model can make for better results. These inputs are called prompts, and the practice of writing them is called prompt engineering.

While it may sound technical, prompt engineering is the main way to communicate with a large language model (LLM), including those that generate video, like the OpeanAI’s Sora. 

Prompt is the way of “Getting what you need” from an AI model.

And this is generating new types of white-collar work. Businesses as disparate as media, consulting, fast-food chains, pharmacies, and automakers—not to mention governments, hospitals, and small—to medium-sized businesses—are employing prompt engineers.

Why Do Prompts Matter To A Publishing Professional?

Because communicating with LLMs doesn’t require a coding background, editorial employees are also part of the prompt engineering process and can work in tandem with data scientists and engineers.

Buzzfeed, for example, experimented with prompts early on and even integrated this technology into Slack, enabling employees to collaborate on prompts and witness how others were using the AI tools.

Publishers aren’t hiring roles specifically for this task yet. Journalists are inherently good at asking questions, so it makes sense that they become part of the prompt design. 

And because a good prompt could differentiate your content or chat experience from all the others, publishers are secretive about their prompts. It is the way they put their fingerprint on what users consume. 

Because a good prompt differentiates content from others, publishers are secretive about the prompts they use

Prompt engineering is necessary for most interactions with LLMs, especially for publishers looking to repurpose and increase the reach of text-based articles through chatbots, quizzes, and, more particularly, videos.

Tips For Good Prompt Writing

A well-crafted prompt must typically include the following:

  • Instruction: these are the primary questions or tasks you’re asking the model to perform.

  • Context: Additional background information that helps the model understand the query better.

  • Inputs: Data or parameters that the model might need to generate the desired output.

  • Examples: Illustrative scenarios that can guide the model towards the type of answer you’re looking for.

  • Output: Examples of what you want the model to generate and in what format.

 

Top 10 Prompts For Publishing Professionals

1. "Write a [adjective] [platform] post about [topic]."

Use this prompt to generate content ideas tailored for different platforms. Replace '[adjective]' with a descriptor like 'insightful' or 'provocative,' choose a relevant platform such as 'blog' or 'LinkedIn,' and insert a pertinent publishing topic. 

This prompt structure helps AI models generate more focused and appropriate content that aligns with the intended audience.

2. "Voice and style guide: Write at a [grade/degree] level. Use [clear/simple/difficult/elaborate] language. Bias toward short sentences. Avoid [jargon, acronyms, etc.]

This prompt guides AI models in adjusting complexity and clarity for targeted audiences by specifying the educational level ('grade' or 'degree') and choosing the language style ('clear,' 'simple,' 'difficult,' and 'elaborate'). 

It emphasizes brevity in sentence structure and directs to avoidance of jargon and acronyms to enhance readability and accessibility.

3. "Write [number] headlines for an article on [topic] for [audience]."

While self-explanatory, this kind of prompt can speed up workflow. It helps publishers develop content for specific target audiences, or A/B test different headlines to gauge engagement and effectiveness. 

It's also ideal to explore various angles and messaging strategies during the brainstorming phase that resonate with the intended readership. The approach can enhance the appeal and reach of articles across diverse platforms.

4. "Using the above, write [number] attention-grabbing and creative titles for [topic name], with the keyword [keyword(s)] included in the title."

These prompt formats are most effective when focusing on SEO or capturing audience interest for specific content. 

It's particularly useful when ranking higher in search engine results or targeting keywords for a marketing campaign. Employing this prompt helps ensure that titles are creative and optimized for visibility and relevance to the desired search terms.

However, it is important to note that AI models may not consider factors like keyword stuffing, which may lead to a violation of Google Quality Guidelines.

5. "Generate a blog post outline on the topic [add topic]. The outline must have [X] subheadings."

These prompts are useful for publishers during the initial stages of content creation, especially when organizing complex information or long-form content. They help structure the information faster, making it easier to write and more digestible for readers. 

They're ideal for creating detailed, well-organized articles that cover a topic thoroughly under specified subheadings for enhancing readability and engagement.

6. "Suggest some H2 and H3 subheadings for the content given below. [include content]"

This is suited for refining and structuring existing content to improve readability and SEO. It helps break complex information into manageable sections, making the content more engaging and easier for readers to navigate. 

It is ideal for enhancing the organization of articles and ensuring that key points are effectively highlighted through appropriate subheadings.

7. "Create a blog content calendar for the month of [specific month and year] targeting [your target audience or niche]. Each week should have a distinct theme or focus."

This prompt helps publishers maintain a consistent content pipeline. It helps plan and organize content in advance, ensuring a coherent flow of topics that cater to a target audience or niche. 

AI models use such prompts to build engagement through thematic consistency and anticipation, making managing content throughout the month easier.

8. "Write a blog post introduction on [Topic]. Write it in a [X] tone. Use transition words."

These prompts are useful when developing new content, particularly to keep readers engaged from the start. It tells the AI model to ensure the introduction is attention-grabbing and smoothly transitions into the main content. This helps maintain reader interest and improve the flow and coherence of the article.

9. "Recommend some design ideas for my event pop-up display."

This prompt is ideal for publishers looking to enhance their event displays' visual appeal and effectiveness. This can be useful for trade shows, exhibitions, or promotional events where catching the eye of attendees is crucial. 

This helps brainstorm creative and practical design solutions that reflect the event's theme, attract attention, and convey key messages effectively to the target audience.

10. “Write an article [topic of the desired article], in the style of my previous article on this topic [paste about 250-word sample].”

This prompt helps publishers maintain consistency in voice and style across articles, ensuring a cohesive reader experience. Referencing a previous piece guides the AI model to align new content with established tones and structures, enhancing brand identity.

Common Prompt Engineering Techniques

Advanced prompting engineering techniques can achieve more complex tasks and improve the reliability and performance of LLMs. Publishers must be aware of basic prompt engineering techniques:

Type

What is it?

When to use?

Zero-shot Prompt

The AI model is not provided with any examples or context to help understand the task. It relies on general knowledge and the ability to interpret the prompt.

Useful for quick access to information like definitions or specific questions.

One-shot Prompt

This prompt provides the AI with a single example to demonstrate the desired task, aiding the model in understanding any pattern or format requirements.

For example, formatting headlines for consistency across different articles.

Few-shot Prompt

Similar to a one-shot prompt, but provides multiple examples to help the AI better understand the desired output, allowing the model to generalize the task more effectively.

To summarize news articles focusing on maintaining neutrality and covering key factual details.

Chain-of-
Thought Prompt

It involves a series of connected questions, with the model's responses to previous prompts influencing its understanding and answers to subsequent prompts. Helps maintain context and deepen topic exploration.

Ideal for complex tasks such as creating a detailed and interconnected article or series of articles that build upon one another (coverage of an ongoing story or an election campaign)

How will Gen AI affect Publishers?

Research by McKinsey suggests that AI could add up to $4.4 trillion annually to the global economy across almost every sector, from banking to media to life sciences. One of Gen AI’s strengths is that it can help nearly everyone with their jobs.

Knowing prompt engineering will help publishers use Gen AI tools effectively to identify and synthesize trends, key drivers, and market and product opportunities from vast unstructured data such as social media, news, academic research, and user feedback.

To maximize effectiveness, publishers must maintain a detailed log of tested prompts, sample outputs, and a scoring rubric. This will help assess which prompts yield the best results.

Want to learn more about prompt engineering? Here is a detailed guide.