To use ChatGPT or not to use ChatGPT? Is that really the question anymore?<\/p>\n
<\/a><\/p>\n If you’re like me, you already use this tool to amplify your awesomeness. Like improving your email writing, analyzing data faster, or firing up your social media strategy.<\/p>\n But I have to admit \u2014 it\u2019s not all peaches and cream. And others seem to agree \u2014 marketers we surveyed stated ChatGPT outputs aren\u2019t publishable 96% of the time<\/a>.<\/p>\n Does this mean you should throw the whole AI concept away? Not at all. Your outputs are only as good as your prompting skills.<\/p>\n So, I spoke to HubSpot’s SEO team to understand how to write prompts for ChatGPT. Stick around to see the key elements of a good prompt, helpful ChatGPT prompt tips, and prompt writing examples.<\/p>\n Table of Contents<\/strong><\/p>\n Prompt Writing Techniques<\/a> <\/a> <\/p>\n It’s tempting to load up ChatGPT and jump right into the prompt. However, Josh Blyskal<\/a>, Associate SEO Strategist at HubSpot, recommends a more introspective approach.<\/p>\n \u201cThe first thing you should do is figure out what you want. Before I type anything, I always take a moment to envision what the final output will look like,\u201d he told me.<\/p>\n He gives an example of a marketer who wants to generate a marketing plan.<\/p>\n Instead of typing \u201cWrite me a marketing plan\u201d into ChatGPT, she should take the time to envision what the plan looks like \u2014 how many sections it has, how many bullet points are under each section, and what information it contains.<\/p>\n Now that she has a better grasp of the task at hand, it becomes much easier to compose a prompt that checks all the boxes.<\/p>\n AI models thrive on specificity. If you feed it with vague instructions, you give it room to interpret what it thinks<\/em> you want.<\/p>\n For example, suppose you prompt ChatGPT to \u201cRewrite this [article\/email\/Tweet],\u201d expecting it to write something new and original. Instead, it regurgitates your original message using the same words and phrases.<\/p>\n The solution is to get specific and intentional with your words.<\/p>\n Instead of \u201cRewrite,\u201d for instance, you might say reimagine<\/em>, expand<\/em>, simplify<\/em>, or modernize<\/em>. Even a small change in the way you phrase a prompt can significantly change the response.<\/p>\n As Blyskal points out, \u201cA lot of people treat AI like a rigid system that can be hacked with the right words or phrases, but that’s not the case.<\/p>\n There are no secret keywords that can guarantee the best results. It’s like playing a game of words, and you have to think intentionally about the ones you use.\u201d<\/p>\n You could get away with stopping at the previous step \u2014 especially if your request is fairly straightforward (e.g., \u201cWrite five knock-knock jokes for kids\u201d).<\/p>\n However, you can supercharge your prompts by padding them with context. This gives the AI model more ingredients to cook up a great response.<\/p>\n Bianca D’Agostino<\/a>, former SEO Specialist at HubSpot, puts it this way:<\/p>\n \u201cSuppose your friend asks you to go to the store and pick up a few things. You definitely need more context \u2014 which store, what things, what’s the budget, and so on. The more context you get from your friend, the easier it is to do the task. The same goes for AI!\u201d<\/p>\n Let’s look at an example:<\/p>\n\n Both of these prompts might yield good results, but the second prompt will give you a more detailed and compelling description.<\/p>\n This is because the AI model has more context to work with, including the product’s name (\u201cSqueaky Queen\u201d), its form (a scrub), and its key features (vegan, gluten-free, recyclable packaging).<\/p>\n Although AI models are powerful, they still need you to take the lead. Without clear instructions, they’re essentially a chef without a recipe or a new employee without a handbook.<\/p>\n Here’s an example: Imagine you need to write an article about Instagram marketing. Along with the main task (“Write an article about Instagram marketing<\/em>“), you also instruct ChatGPT to:<\/p>\n \u201cCover the following subtopics\u2026\u201d Not only are you giving ChatGPT a clear topic, but you\u2018re also guiding and shaping the content itself. You\u2019re adding quotes, statistics, and examples to give the output more texture.<\/p>\n Of course, it’s not just what you say but how you say it. The content of the article might be up to par, but it may lack a unique voice or perspective. This is where style and tones come in.<\/p>\n For Sylviain Charbit<\/a>, HubSpot Senior Technical SEO Specialist, adding style to prompts is a multi-step process. He starts by letting the AI tool generate a general answer and then progressively molds its response for style.<\/p>\n \u201cYou can ask it to inspire itself from a situation, character, or profession, or provide it with already made examples to emulate,\u201d he told me.<\/p>\n Charbit also incorporates ChatGPT tone modifiers, such as \u201ccheerful,\u201d \u201chumorous,\u201d or \u201cauthoritative,\u201d to guide the AI further.<\/p>\n Let’s look at some examples:<\/p>\n\n Other style parameters include length and format. You can specify the length of the response (e.g., \u201cKeep the article under 800 characters\u201d) and its format (e.g., \u201cCreate a list of bullet points\u201d).<\/p>\n Blyskal follows the 60% Rule.<\/p>\n He told me, \u201cWhen you generate something and it’s at least 60% of what you want, you should continue refining that specific prompt. But if it falls short of that, it might be time to start from scratch.\u201d<\/p>\n So, if you ask ChatGPT to generate a list of ten blog ideas and you only like two, it’s a good indication to go back to the drawing board.<\/p>\n Even if you meet the 60% threshold, you can still refine the output by giving follow-up instructions such as, \u201cMake it sound more conversational\u201d or \u201cCondense this into one paragraph.\u201d<\/p>\n Remember that prompt writing isn’t an exact science. Like any creative process, it often takes multiple revisions to get the right output.<\/p>\n <\/a> <\/p>\n Role-playing is a technique that involves asking the AI model to adopt a specific role, job, or function.<\/p>\n \u201cWhen you ask an AI model to play a specific role, you’re giving it a certain level of expertise,\u201d Blyskal told me. \u201cFor instance, if you prompt it to act as a professional interviewer, it will look for examples of good interviewing skills and respond with professional-seeming questions.\u201d<\/p>\n By defining a role for the AI model to play \u2014 such as a professional interviewer, a graphic designer, or a nutritionist \u2014 you’re also tapping into different viewpoints and perspectives, which can be especially helpful for marketers.<\/p>\n For example, let\u2018s say you\u2019re launching a new product, a cold plunge tub. You need to create an ad that appeals to your target audience: long-distance runners. In this scenario, you ask the AI model to play the role of a long-distance runner and list the key benefits of your product.<\/p>\n Here\u2019s an example from Perplexity (my fave AI tool) because it provides images and links to sources (peace out hallucinations):<\/p>\n\n Now, the AI model will generate responses that align with the needs and preferences of long-distance runners. This gives you valuable insight into your audience, enabling you to create a more compelling and relevant ad.<\/p>\n Imagine you\u2018re on a road trip, and you\u2019re relying on GPS to guide you from point A to point B. Throughout the journey, it gives you turn-by-turn directions until you reach your destination.<\/p>\n Chain of thought prompting follows a similar approach. You give the AI model one prompt after another, with each prompt building on the one before it. This keeps the model on track, giving you greater control of the output.<\/p>\n\n \u201cYou’re basically asking the AI model to explain its reasoning at each step. If you don’t like any of the explanations, you can tinker with the output until it’s ready for the next prompt,\u201d Blyskal explains.<\/p>\n This is ideal for tasks that can be broken down into smaller steps.<\/p>\n For example, consider a copywriter using AI to create an article. She first prompts it to create an outline. Once she\u2018s happy with the outline, she prompts it to write the article, and then the headline, and then the meta description. She doesn\u2019t move to the next step until she’s satisfied with the current output.<\/p>\n That’s the power of chain prompting. By guiding the model in a logical sequence, you generate more accurate and relevant responses.<\/p>\n \u201cIf there’s any one tip that will take your prompting to the next level, it’s providing examples to the AI model,\u201d Blyskal told me.<\/p>\n You can think of examples as reference points, helping the AI model emulate a certain style, tone, structure, or format. On top of that, AI can analyze a piece of content and offer improvements.<\/p>\n Here are several types of examples you can provide to ChatGPT:<\/p>\n \u00a0Style and tone reference: <\/strong>Don\u2019t ask for a blog post in a specific topic and hope for the best \u2014 provide a sample paragraph or excerpt showing the style and tone you want. For instance, include a snippet of witty and conversational text, if that\u2019s what you\u2019re going for. <\/p>\n \u00a0Structural guidance: <\/strong>Include a sample or visual outline when requesting a content outline. It should show the expected flow, sections, and key content elements (e.g., intro, X subheadings, bullet lists, etc.). <\/p>\n \u00a0Content improvement suggestions: <\/strong>After drafting content with ChatGPT, provide specific examples of areas that need improvement or refinement. Highlight sentences or sections that need clarification, enhancement, or a different approach. Don\u2019t forget to include style and tone references to ensure it stays consistent. <\/p>\n While AI prompting is a science, it\u2019s also an art. So don\u2019t be afraid to get creative with your examples and instructions to get out-of-the-box outputs.<\/p>\n <\/a> <\/p>\n Here are three use cases I\u2019m seeing many people use in content marketing today. I included the prompt, why it works, and the generated output by ChatGPT.<\/p>\n Prompt: <\/strong>\u201cCreate a social media campaign to launch our new eco-friendly water bottle. The campaign should include 5 Instagram posts, 3 Facebook posts, and 2 X posts. Each post should highlight the product’s sustainability features, target eco-conscious consumers, and include a call-to-action to visit our website.\u201d<\/p>\n Why it works:<\/strong><\/p>\n Context:<\/strong> Specifies the product and its unique selling point (eco-friendly water bottle). <\/p>\n Task:<\/strong> Clearly defines the number and type of social media posts needed. <\/p>\n Instructions:<\/strong> Details what each post should highlight (sustainability features) and the target audience (eco-conscious consumers). <\/p>\n Clarification:<\/strong> Includes a call-to-action to drive traffic to the website. <\/p>\n
\n How to Write the Perfect AI Prompt<\/a>
\n ChatGPT Prompt Writing Examples<\/a> <\/p>\nHow to Write the Perfect AI Prompt<\/strong><\/h2>\n\n
1. Envision your ideal response.<\/strong><\/h3>\n
2. Write a specific, action-oriented task.<\/strong><\/h3>\n
3. Set the stage with context.<\/strong><\/h3>\n
4. Add clear instructions.<\/strong><\/h3>\n
\n \u201cIncorporate the following quote\/statistic\u2026\u201d
\n \u201cAdd an example to demonstrate\u2026\u201d <\/p>\nAdding Style and Tone<\/strong><\/h4>\n
5. Clarify and refine.<\/strong><\/h3>\n
Prompt Writing Techniques<\/strong><\/h2>\n
Give It a Role<\/strong><\/h3>\n
Chain of Thought<\/strong><\/h3>\n
Provide Examples<\/strong><\/h3>\n
ChatGPT Prompt Writing Examples<\/strong><\/h2>\n
Social Media Campaign for a New Product Launch<\/strong><\/h3>\n