Tired of getting bland, generic responses from AI tools? In this episode I break down the art of effective prompting - think of it as learning how to give AI the perfect creative brief so you actually get what you want. No tech background required, just practical techniques that'll transform your AI interactions from frustrating to fantastic.
🎁 BONUS: Get all the copy-paste ready prompt templates and examples from this episode - CLICK HERE.
Feel free to duplicate this page and save it in your Notion workspace for easy access!
KEY POINTS- The quality of your prompt dramatically affects your results - same tool, wildly different outputs based on how you ask
- Seven core principles for better prompting: be specific, provide context, specify format, include examples, use role prompting, iterate, and set system prompts
- Different creative disciplines need different prompting approaches - frameworks tailored for writers, designers, marketers, photographers, and more
- AI doesn't understand implied context like humans do - you need to spell out everything you'd normally assume
- Effective prompting is about iteration, not getting it perfect the first time
- Examples and references work like magic - show AI what you want, don't just describe it
- Role prompting transforms responses - ask AI to adopt specific professional perspectives for dramatically better results
- CATP Framework (for writers): Context, Audience, Tone, Purpose
- VCRT Approach (for designers): Visual, Context, Reference, Technical
- PAISM Method (for marketers): Problem, Audience, Insight, Solution, Measurement
- System Prompts: Setting context for extended AI conversations
- Role Prompting: Having AI adopt specific professional perspectives
- Try the before/after test - Take a vague prompt you've used and rewrite it with specific details, context, and examples
- Pick one framework from this episode that matches your creative field and experiment with it on your next project
- Practice iteration - Don't accept the first AI response; refine your prompt based on what you get back
- Include examples in your prompts - show AI what good looks like rather than just describing it
- Set up a system prompt for your next extended AI conversation to maintain consistency
00:00:00 - Welcome back! Why prompting makes or breaks your AI experience
00:03:06 - What is prompting and why does it matter so much?
00:07:03 - The Seven Universal Laws of Effective Prompting
00:16:00 - Discipline-specific frameworks: CATP framework for writers
00:18:40 - VCRT approach for graphic and UI/UX designers
00:20:50 - PAISM framework for marketers and strategists
00:22:17 - Techniques for photographers and videographers
00:23:21 - Podcast and audio creator prompting strategies
00:25:38 - Seven common prompting mistakes (and how to avoid them)
00:31:25 - Recap and teaser for next week's advanced techniques
COMING NEXT WEEKPart two of our prompting series! We'll explore advanced techniques like chain-of-thought prompting for complex problems, reverse prompting (getting AI to help you write better prompts), comparative prompting for exploring multiple solutions, and more. Plus, a real-world demonstration of the entire prompt crafting and refinement process.
Ready to transform your AI interactions? Download the complete framework templates from this episode HERE and never struggle with prompting again!
Let's Connect!00:00:00 --> 00:00:04 Are you still doing the creative tasks that AI could handle in a fraction of the time?
00:00:04 --> 00:00:10 I used to spend four hours on tasks that now take me 30 minutes with better results.
00:00:10 --> 00:00:13 That's not just efficiency, it's creative freedom.
00:00:13 --> 00:00:16 I'm Tiana and this is AI for Creative Professionals.
00:00:17 --> 00:00:33 Music.
00:00:33 --> 00:00:38 Hey there, welcome back to AI for Creative Professionals, the podcast that helps
00:00:38 --> 00:00:43 designers, writers, marketers, and all kinds of creators navigate this wild
00:00:43 --> 00:00:45 world of artificial intelligence.
00:00:45 --> 00:00:48 I'm Tiana, and I'm thrilled you're joining me for episode two.
00:00:49 --> 00:00:54 So last week, we covered the basics of what AI is, how we got here,
00:00:54 --> 00:00:57 and some first steps for getting started.
00:00:57 --> 00:01:00 If you missed that episode, definitely go back and check it out.
00:01:00 --> 00:01:03 But don't worry, you can still follow along with today's episode just fine.
00:01:04 --> 00:01:07 I'm sure you've seen this all over social media.
00:01:07 --> 00:01:15 And it's the fact that there's quite a bit of difference between getting somewhat
00:01:15 --> 00:01:22 useless generic results from AI and getting amazing,
00:01:22 --> 00:01:28 useful results that are actually going to improve your workflow.
00:01:28 --> 00:01:35 So that's what we are going to focus on today, the art of effective prompting.
00:01:35 --> 00:01:40 And by the end of this episode, you'll understand what makes a good prompt.
00:01:40 --> 00:01:44 How to structure your requests to get better results, and
00:01:44 --> 00:01:50 specific techniques you can use that are tailored for different creative disciplines
00:01:50 --> 00:01:56 and I'll share some examples and common mistakes to avoid and also the frameworks
00:01:56 --> 00:02:02 you can start using immediately to improve your AI interactions.
00:02:02 --> 00:02:09 Honestly, this might be one of the most valuable skills you can develop for working with AI tools.
00:02:10 --> 00:02:15 It's like learning how to write a good creative brief or how to communicate
00:02:15 --> 00:02:21 effectively with clients and collaborators. except now you're communicating with an AI.
00:02:22 --> 00:02:28 So let's dive in and discover some techniques that are definitely going to improve
00:02:28 --> 00:02:35 your communication with AI and deliver exactly the type of results that you need.
00:02:36 --> 00:02:43 All right, so what exactly is prompting and why am I dedicating an entire episode to it?
00:02:43 --> 00:02:49 Simply put, a prompt is the instruction or question you give to an AI tool like
00:02:49 --> 00:02:54 ChatGPT or Claude. It is how you communicate what you want.
00:02:55 --> 00:03:00 And the quality of your prompt dramatically affects the quality of what you get back.
00:03:01 --> 00:03:04 It's your interface with the AI's capabilities.
00:03:04 --> 00:03:10 The better you prompt, the better your results. Let's picture a story that illustrates
00:03:10 --> 00:03:12 how much prompting matters.
00:03:12 --> 00:03:16 Let's say you're writing blog posts about sustainable packaging.
00:03:16 --> 00:03:23 And you're in a rush. you go to ChatGPT and you type, write an article about
00:03:23 --> 00:03:24 sustainable packaging.
00:03:24 --> 00:03:31 And what comes back is, well, generic, broad, something you could find on thousands
00:03:31 --> 00:03:33 of different websites, right?
00:03:33 --> 00:03:36 It definitely hits all the obvious points. Plastic pollution,
00:03:36 --> 00:03:40 recyclable materials, the usual suspects, nothing wrong with it,
00:03:40 --> 00:03:42 but nothing special either.
00:03:42 --> 00:03:48 So there's a different approach that you could use, instead of that very simple
00:03:48 --> 00:03:54 request, you can type, write a comprehensive guide to sustainable packaging
00:03:54 --> 00:03:56 innovations for small food businesses.
00:03:56 --> 00:04:03 Focus on solutions that are both eco-friendly and cost-effective for companies with limited budgets.
00:04:03 --> 00:04:09 Include specific material recommendations, their environmental impact,
00:04:09 --> 00:04:12 approximate costs, and where to source them.
00:04:12 --> 00:04:17 The article should be informative but conversational, aimed at food entrepreneurs
00:04:17 --> 00:04:22 who are passionate about sustainability but concerned about costs.
00:04:23 --> 00:04:27 Suddenly, you will have a specific actionable content with real value,
00:04:28 --> 00:04:32 and it will discuss particular materials with price points, offer practical
00:04:32 --> 00:04:38 advice for small businesses, and have much more of an engaging tone.
00:04:38 --> 00:04:40 And it all comes from the same AI tool.
00:04:40 --> 00:04:47 The only difference was how we asked, right? And I've seen this pattern repeat over and over.
00:04:47 --> 00:04:52 The real limitation often isn't the AI's capabilities, but rather our ability
00:04:52 --> 00:04:55 to effectively communicate what we want.
00:04:55 --> 00:05:00 It's like the difference between a vague client who says, make it pop,
00:05:01 --> 00:05:05 right? and a client who provides a detailed, thoughtful brief.
00:05:05 --> 00:05:09 The end results reflect the quality of that initial direction.
00:05:10 --> 00:05:12 This is specifically true for creative professionals.
00:05:13 --> 00:05:18 Our work often requires nuance, a specific aesthetic directions.
00:05:18 --> 00:05:26 Tonal considerations, and contextual understanding. We can't just ask for a logo or a website copy.
00:05:27 --> 00:05:31 We need to provide context, constraints, and clear direction.
00:05:31 --> 00:05:36 Let's say you're a photographer and you type into AI image generation tool things
00:05:36 --> 00:05:41 like, give me an image of beautiful landscape, and obviously you're going to
00:05:41 --> 00:05:44 get generic, boring results.
00:05:44 --> 00:05:50 But if you craft a detailed prompt with specific lighting conditions,
00:05:51 --> 00:05:57 compositional details, and stylistic references, the output will transform into
00:05:57 --> 00:06:02 genuinely useful reference images for your photoshoot, right?
00:06:03 --> 00:06:08 So here's the key insight that changes how you should approach prompting.
00:06:09 --> 00:06:14 AI models don't understand your intent the way a human would.
00:06:14 --> 00:06:20 They don't have the benefit of all the unstated context, assumptions,
00:06:20 --> 00:06:23 and shared references that happen in human communication.
00:06:24 --> 00:06:28 Think about it. When you brief a colleague on a project, there's so much that
00:06:28 --> 00:06:34 goes unsaid. You have shared cultural references, industry knowledge,
00:06:34 --> 00:06:38 previous conversations you had, understanding of the client.
00:06:38 --> 00:06:44 But with AI, you need to spell out those assumptions and contexts that would
00:06:44 --> 00:06:47 be automatic with a human collaborator.
00:06:47 --> 00:06:52 The good news is that effective prompting is a learnable skill.
00:06:52 --> 00:06:57 And once you master it, it applies across pretty much all AI tools.
00:06:57 --> 00:07:02 Let's get into the nuts and bolts of how to craft prompts that actually work.
00:07:04 --> 00:07:09 Let's talk about some fundamental principles that apply across all AI tools,
00:07:09 --> 00:07:15 whether you're using them for writing, image generation, or any other creative application.
00:07:16 --> 00:07:20 I think of these as the universal laws of good prompting.
00:07:20 --> 00:07:27 Once you internalize these, you'll start getting dramatically better results from any AI tool you use.
00:07:27 --> 00:07:31 So, principle number one is be specific and detailed.
00:07:32 --> 00:07:36 Big prompts lead to vague results. The more specific you are about what you
00:07:36 --> 00:07:40 want, the more likely you're going to get it.
00:07:40 --> 00:07:43 Let me give you a before and after example.
00:07:43 --> 00:07:47 Before. Write a product description for a coffee maker.
00:07:47 --> 00:07:53 After. Write a premium product description for a high-end, minimalist coffee
00:07:53 --> 00:07:56 maker designed for pour-over brewing.
00:07:56 --> 00:08:03 The target audience is design-conscious urban professionals who appreciate artisanal coffee.
00:08:04 --> 00:08:09 The tone should be sophisticated yet approachable, emphasizing craftsmanship,
00:08:09 --> 00:08:11 precision, and elegant design.
00:08:11 --> 00:08:16 Include details about materials, walnut wood and borosilicate glass,
00:08:17 --> 00:08:23 brewing capability, 500 milliliters, and the hand-crafted nature of the product.
00:08:23 --> 00:08:27 So the first prompt could generate a description for any coffee maker,
00:08:27 --> 00:08:32 from a basic $20 drip machine to a professional espresso system.
00:08:32 --> 00:08:36 The second gives the AI clear parameters about product type,
00:08:37 --> 00:08:41 audience, tone, materials, and key selling points.
00:08:42 --> 00:08:46 Principle number two, provide context and purpose.
00:08:46 --> 00:08:51 AI tools don't know if you're asking for something or how you plan to use it, right?
00:08:51 --> 00:08:57 When you provide this context, the AI can tailor the response appropriately.
00:08:57 --> 00:09:02 For example, instead of give me ideas for a social media campaign,
00:09:02 --> 00:09:08 you can try, I'm developing a social media campaign for a local bookstore trying
00:09:08 --> 00:09:13 to compete with Amazon and increase food traffic to their physical location.
00:09:13 --> 00:09:19 Their key differentiators are personalized recommendations, community events,
00:09:19 --> 00:09:21 and supporting local authors.
00:09:21 --> 00:09:28 Generate five creative campaign ideas that emphasize these unique aspects and
00:09:28 --> 00:09:32 encourage people to visit the store rather than ordering online.
00:09:33 --> 00:09:38 Now the AI understands the business challenge, the competitive situation,
00:09:38 --> 00:09:42 the key selling points, and the specific objective of the campaign.
00:09:43 --> 00:09:51 The context helps AI understand not just what you want, but why and for whom,
00:09:51 --> 00:09:54 resulting in much more relevant suggestions.
00:09:54 --> 00:09:58 Principle number three, specify format and structure.
00:09:59 --> 00:10:05 Don't make AI guess how you want information presented. Be explicit about the
00:10:05 --> 00:10:08 format, length, and structure you prefer.
00:10:09 --> 00:10:12 Instead, tell me about email marketing best practices.
00:10:13 --> 00:10:18 You can try. Create a structured guide to email marketing best practices for
00:10:18 --> 00:10:20 a small creative agency.
00:10:20 --> 00:10:25 Format it as bulleted lists organized into five categories.
00:10:25 --> 00:10:32 Subject lines, content structure, visual elements, call-to-action strategies, and testing methods.
00:10:32 --> 00:10:38 For each category, include three to four actionable tips with brief explanations.
00:10:38 --> 00:10:44 Keep the entire guide under 800 words and use a professional but conversational tone.
00:10:45 --> 00:10:51 This gives the AI clear instructions about how to organize and present the information,
00:10:51 --> 00:10:54 which almost always results in more useful outputs.
00:10:54 --> 00:10:59 I use this technique all the time for my podcast clients when creating show notes.
00:10:59 --> 00:11:06 Instead of just saying to AI, write a show notes for this specific episode and
00:11:06 --> 00:11:12 then providing it with a transcript, I specify exactly what sections I want, like summary,
00:11:12 --> 00:11:16 key points, timestamps, resources mentioned, call to action,
00:11:16 --> 00:11:19 and how long each section should be.
00:11:19 --> 00:11:21 It saves me so much editing time.
00:11:22 --> 00:11:25 Principle number four, include examples or references.
00:11:26 --> 00:11:33 This is a lifesaver because if you have examples of what you're looking for, please include them.
00:11:33 --> 00:11:38 This gives the AI a much clearer idea of your expectations and preferences.
00:11:39 --> 00:11:42 For instance, if you're looking for a certain writing style,
00:11:42 --> 00:11:48 you might say, I'd like you to write in a style similar to this example,
00:11:48 --> 00:11:51 and then you would insert an example paragraph.
00:11:51 --> 00:11:57 Using this style, write a short blog introduction about the benefits of meditation
00:11:57 --> 00:11:58 for creative professionals.
00:11:59 --> 00:12:03 Principle number five, use role and perspective prompting.
00:12:03 --> 00:12:07 So this is a technique that I'm sure you've seen all over Instagram.
00:12:08 --> 00:12:14 And it is the technique where you ask the AI to adopt a specific role or perspective,
00:12:14 --> 00:12:17 which can dramatically change how it approaches your requests.
00:12:17 --> 00:12:22 Instead of saying, give me feedback on this logo design, you can try.
00:12:22 --> 00:12:27 Take on the role of an experienced brand designer who specializes in retail
00:12:27 --> 00:12:33 brands. Review this logo concept for a premium organic grocery store called Harvest Market.
00:12:34 --> 00:12:37 Provide constructive feedback from your professional perspective,
00:12:37 --> 00:12:44 focusing on visual hierarchy, color psychology, scalability, and brand alignment.
00:12:45 --> 00:12:49 Include both strengths and areas for improvement.
00:12:49 --> 00:12:54 So this approach leverages the AI's training on how different professionals
00:12:54 --> 00:12:57 with different expertise might approach a problem.
00:12:57 --> 00:13:02 I use this technique when I want to get unstuck creatively.
00:13:02 --> 00:13:09 Recently, I was working on a difficult messaging project and I felt like I was just going in circles.
00:13:10 --> 00:13:16 And I asked Claude to take on the perspective of a veteran copywriter who specializes
00:13:16 --> 00:13:21 in distilling complex value positions into simple, compelling messages.
00:13:21 --> 00:13:26 And this shift in perspective introduced new ideas, for sure.
00:13:26 --> 00:13:33 Principle number six, iterate and refine. This might be the most important principle of all.
00:13:33 --> 00:13:36 Don't expect perfection in one go.
00:13:36 --> 00:13:42 Treat prompting as an iterative process where you can refine based on initial results.
00:13:42 --> 00:13:48 Start with the prompt, see what you get, and then clarify or adjust based on the response.
00:13:48 --> 00:13:52 You might say, that's a good start, but the tone is too formal.
00:13:52 --> 00:13:58 Can you revise it to be more conversational and engaging like you're explaining to a friend?
00:13:59 --> 00:14:04 Or, I left structure, but the examples aren't relevant to my target audience
00:14:04 --> 00:14:06 of small business owners.
00:14:06 --> 00:14:11 Can you provide examples specifically relevant to solo entrepreneurs and small
00:14:11 --> 00:14:13 businesses with limited budgets?
00:14:14 --> 00:14:19 Principle number seven, use system prompts for consistent interactions.
00:14:19 --> 00:14:24 So if you're going to have an extended interaction with an AI.
00:14:25 --> 00:14:31 Setting expectations up front with a system prompt can be incredibly helpful.
00:14:31 --> 00:14:37 A system prompt is what establishes the context for the entire conversations.
00:14:37 --> 00:14:42 For example, I'm a graphic designer working on brand for a new eco-friendly
00:14:42 --> 00:14:48 children's soy company. In our conversation, I'll be asking you for ideas and
00:14:48 --> 00:14:51 feedback on various aspects of the brand.
00:14:51 --> 00:14:56 Please keep your responses focused on sustainable, child-friendly design principles
00:14:56 --> 00:15:02 and feel free to ask clarifying questions if my requests are vague.
00:15:02 --> 00:15:07 I prefer specific, actionable suggestions rather than generic principles.
00:15:07 --> 00:15:14 And this sets the stage for all subsequent prompts and creates a more consistent, helpful interaction.
00:15:14 --> 00:15:19 I use System Prompts extensively when I'm working on long projects.
00:15:19 --> 00:15:25 For instance, when I'm developing a content plan for clients' Instagram content,
00:15:25 --> 00:15:32 I'll start by telling Claude all about the niche, topic, audience, goals, and tone.
00:15:32 --> 00:15:37 And then throughout our conversations, it has that context and it can provide
00:15:37 --> 00:15:39 me with more relevant suggestions.
00:15:40 --> 00:15:46 These seven principles form the foundation of effective prompting across all AI tools.
00:15:47 --> 00:15:51 Master these and you'll immediately start getting better results from any AI
00:15:51 --> 00:15:53 assistant you work with.
00:15:53 --> 00:15:58 Now, let's look at how to apply these principles to specific creative disciplines.
00:15:59 --> 00:16:06 Let's get into practical and effective prompting for various creative disciplines.
00:16:06 --> 00:16:11 I'll share specific templates and examples you can adapt for your own work,
00:16:11 --> 00:16:16 whether you're a writer, designer, marketer, or another type of creative professional.
00:16:16 --> 00:16:21 For writers and content creators. One of the most effective frameworks I've
00:16:21 --> 00:16:29 found for writing prompts is what I call the Context Audience Tome Purpose, or CATP framework.
00:16:30 --> 00:16:36 Here's what it looks like in practice. So for context, this is for a blog post
00:16:36 --> 00:16:40 on a financial advisory firm's website. Audience.
00:16:41 --> 00:16:43 Young professionals 25 to 35 who
00:16:43 --> 00:16:48 are beginning to build wealth but feel intimidated by financial planning.
00:16:48 --> 00:16:54 Tone. Approachable, jargon-free, and encouraging. Not preachy or condescending.
00:16:54 --> 00:17:00 Purpose. to explain the concept of compound interest and motivate readers to
00:17:00 --> 00:17:03 start investing early, even with small amounts.
00:17:03 --> 00:17:10 With this in mind, write an engaging introduction, 250 to 300 words,
00:17:10 --> 00:17:15 that uses a relatable analogy to explain compound interest and hooks the reader
00:17:15 --> 00:17:20 emotionally by addressing their potential fears about investing.
00:17:21 --> 00:17:26 Notice how this prompt clearly establishes all the key elements a writer would
00:17:26 --> 00:17:27 need to know before starting a piece.
00:17:28 --> 00:17:33 I used a similar approach when helping a travel blogger who was experiencing writer's block.
00:17:33 --> 00:17:37 She had all her notes from a trip to Portugal but was struggling to turn them
00:17:37 --> 00:17:38 into an engaging blog post.
00:17:38 --> 00:17:44 And so we created a prompt that specified her casual first-person storytelling
00:17:44 --> 00:17:49 style, her audience of budget-conscious millennials, and her purpose of highlighting
00:17:49 --> 00:17:53 hidden local spots while providing practical tips.
00:17:53 --> 00:17:58 The AI generated an excellent structural outline and opening paragraphs that
00:17:58 --> 00:18:02 maintained her authentic voice while organizing her thoughts.
00:18:03 --> 00:18:08 For content editing, another effective template is review the following text
00:18:08 --> 00:18:11 and improve it in these specific ways. 1.
00:18:11 --> 00:18:16 Make the tone more, and then you would insert your desired tone. 2.
00:18:16 --> 00:18:20 Simplify an overly complex explanation. 3.
00:18:21 --> 00:18:26 Strengthen the call to action. 4. Ensure consistency in terminology throughout.
00:18:27 --> 00:18:30 Here's the text, and then you would paste your content.
00:18:31 --> 00:18:36 This gives the AI specific editing directions rather than a vague request to
00:18:36 --> 00:18:38 quote-unquote make it better, right?
00:18:38 --> 00:18:43 For graphic designers and visual creators, when using AI for concept generation
00:18:43 --> 00:18:49 or reference images, specificity about visual elements is crucial.
00:18:49 --> 00:18:55 A helpful framework is what I call the visual context reference technical or
00:18:55 --> 00:18:59 VCRT approach. So, visual elements.
00:19:00 --> 00:19:06 Create a concept for a logo featuring a stylized fox integrated with geometric patterns.
00:19:06 --> 00:19:10 The fox should be simplified but recognizable, facing right,
00:19:10 --> 00:19:12 with pointed ears and a distinctive tail.
00:19:13 --> 00:19:19 Context. This is for a digital security company called SwiftGuard that wants
00:19:19 --> 00:19:25 to convey intelligence, vigilance, and agility without appearing aggressive. References.
00:19:26 --> 00:19:31 The style should blend aspects of minimal Scandinavian design with subtle tech
00:19:31 --> 00:19:34 elements similar to Cisco's logo.
00:19:34 --> 00:19:40 Color palette should include deep orange and navy blue. Technical considerations.
00:19:41 --> 00:19:47 The logo needs to work in both horizontal and square formats and remain recognizable
00:19:47 --> 00:19:49 when scaled down to fabric in size.
00:19:49 --> 00:19:53 This approach gives the AI clear guidance on visual elements,
00:19:53 --> 00:19:57 brand context, stylistic references, and technical requirements.
00:19:58 --> 00:20:04 For UI UX designers, I recommend including user goals and contexts.
00:20:04 --> 00:20:09 Generate concepts for a mobile app checkout screen with these considerations.
00:20:09 --> 00:20:13 User goal. Complete purchase quickly with minimum steps.
00:20:13 --> 00:20:20 Context Users are often multitasking or in transit when making purchases Design
00:20:20 --> 00:20:27 direction Clean minimal interface with clean visual hierarchy Specific needs
00:20:27 --> 00:20:32 Must include saved payment methods, shipping address selection,
00:20:32 --> 00:20:38 order summary, and prominent checkout button Constraints Limited screen space
00:20:38 --> 00:20:44 on mobile devices Describe three different approaches to organizing these elements
00:20:44 --> 00:20:47 considering user flow and ease of use.
00:20:48 --> 00:20:53 For marketers and strategists. When creating marketing concepts or strategies,
00:20:54 --> 00:21:03 I find problem audience inside solution measurement or P-A-I-S-M framework extremely useful.
00:21:03 --> 00:21:09 Problem. A premium skincare brand is struggling to differentiate itself in a
00:21:09 --> 00:21:14 crowded market where many competitors also claim to be natural and science-backed.
00:21:15 --> 00:21:22 Audience. Professional women 35 to 50 with disposable income who are concerned
00:21:22 --> 00:21:25 about aging but skeptical of marketing claims.
00:21:25 --> 00:21:32 Insight This audience values transparency and education over miraculous claims
00:21:32 --> 00:21:36 and tends to research ingredients thoroughly before purchasing.
00:21:37 --> 00:21:44 Solution Develop a content marketing strategy that emphasizes ingredient education and transparency.
00:21:45 --> 00:21:51 Measurement. Success will be measured by engagement with educational content,
00:21:51 --> 00:21:54 email signups, and first-time purchases.
00:21:54 --> 00:22:00 Based on this brief, create a three-month content marketing plan with specific
00:22:00 --> 00:22:04 content themes, formats, and distribution channels.
00:22:04 --> 00:22:10 This framework ensures your marketing prompts include all the strategic elements
00:22:10 --> 00:22:13 needed for meaningful AI-generated recommendations.
00:22:13 --> 00:22:19 For photographers and videographers, when using AI to generate concepts,
00:22:19 --> 00:22:24 shot lists, or editing approaches, focus on mood, narrative,
00:22:24 --> 00:22:25 and technical specifications.
00:22:25 --> 00:22:31 You can say something like, I'm planning a lifestyle photo shoot for a sustainable
00:22:31 --> 00:22:37 clothing brand targeting outdoor enthusiasts who care about environmental impact.
00:22:37 --> 00:22:42 The goal is to show the clothes in natural settings while conveying a sense
00:22:42 --> 00:22:44 of adventure and environmental consciousness.
00:22:45 --> 00:22:51 Generate a detailed shot list for a half-day shoot at a coastal location,
00:22:51 --> 00:22:56 including three wide environmental shots that establish the setting,
00:22:56 --> 00:23:00 four medium shots showcasing specific garments in action,
00:23:01 --> 00:23:05 three detailed shots highlighting sustainable material features.
00:23:05 --> 00:23:10 Two lifestyle vignettes that tell a story about the brand's ethos.
00:23:10 --> 00:23:16 For each shot, include notes on composition, lightning, subject positioning,
00:23:16 --> 00:23:18 and emotional tone to convey.
00:23:19 --> 00:23:24 For podcasts and audio creators. When developing episode content or structure,
00:23:24 --> 00:23:27 be specific about format, pacing, and tone.
00:23:28 --> 00:23:34 You can say something like, I'm planning a 30-minute podcast episode about creative
00:23:34 --> 00:23:38 burnout for an audience of freelance designers and illustrators.
00:23:38 --> 00:23:45 Help me develop an episode structure that includes a compelling cold open that
00:23:45 --> 00:23:51 captures the feeling of creative burnout, An introduction section explaining why this topic matters.
00:23:52 --> 00:23:55 Three main segments about different aspects of burnout.
00:23:56 --> 00:24:01 Practical recovery strategies section. Closing thoughts and call to action.
00:24:01 --> 00:24:07 For each section, provide specific talking points, potential questions for reflection,
00:24:07 --> 00:24:12 and ideas for maintaining an encouraging rather than lecturing tone.
00:24:12 --> 00:24:16 I personally use this approach for my podcast clients all the time.
00:24:16 --> 00:24:22 Recently, I was helping a business podcast host who felt her episodes were becoming
00:24:22 --> 00:24:25 too interview-centric and passive.
00:24:25 --> 00:24:31 So we used a detailed prompt to redesign her format to include more structured segments,
00:24:31 --> 00:24:37 specific types of questions to create tension and insight, and techniques for
00:24:37 --> 00:24:41 guiding conversations toward actionable takeaways, right?
00:24:41 --> 00:24:46 And the resulting structure gave her episodes more distinctive personality while
00:24:46 --> 00:24:48 maintaining the value of guest expertise.
00:24:49 --> 00:24:53 These discipline-specific frameworks are starting points that you can adapt
00:24:53 --> 00:24:55 to your own particular needs.
00:24:55 --> 00:25:00 The key is to include the elements that matter most for your creative field
00:25:00 --> 00:25:02 and specific project you're working on.
00:25:03 --> 00:25:09 And also remember that effective prompting is about communicating like you would
00:25:09 --> 00:25:14 with a skilled but literal-minded human collaborator who doesn't have access
00:25:14 --> 00:25:16 to all the context in your head.
00:25:16 --> 00:25:21 The more clearly you can articulate what you want, the better results you'll get.
00:25:21 --> 00:25:27 In one of the next episodes, I'll be sharing some advanced techniques that can
00:25:27 --> 00:25:29 take your prompting to the next level.
00:25:29 --> 00:25:35 But for now, let's also talk about some common prompting mistakes and how to avoid them.
00:25:36 --> 00:25:41 So, basically, even with the best techniques, there are some common pitfalls
00:25:41 --> 00:25:44 that can limit the effectiveness of your prompts.
00:25:45 --> 00:25:50 Let me share some of the mistakes I've made and I've seen other people make and how to avoid them.
00:25:50 --> 00:25:53 Mistake number one, being too vague or general.
00:25:54 --> 00:25:58 We've talked about this already, and this is probably the most common issue I see.
00:25:59 --> 00:26:03 Big prompts like, give me some creative ideas for this and this,
00:26:03 --> 00:26:06 or give me logo ideas for this brand.
00:26:06 --> 00:26:09 Simply don't provide enough direction.
00:26:10 --> 00:26:17 Also, here's another thing. When you say to AI, give me ideas for engaging posts,
00:26:17 --> 00:26:24 you have to specify to AI what exactly engaging means to you.
00:26:24 --> 00:26:30 Whether that means both that prompt comments and shares rather than likes,
00:26:30 --> 00:26:34 also for which platform, and what is the business goal.
00:26:35 --> 00:26:40 So the fix for this mistake is to always ask yourself if the prompt contains
00:26:40 --> 00:26:48 specific details about format, tone, purpose, audience, and constraints, and if not, add them.
00:26:48 --> 00:26:52 Mistake number two, overloading with too many requirements.
00:26:52 --> 00:26:56 So here is the opposite problem, right?
00:26:56 --> 00:27:00 You're trying to cram too many specific requirements into one prompt,
00:27:00 --> 00:27:05 which just creates a confusing and contradictory set of instructions.
00:27:05 --> 00:27:11 Let's say you're working on a website copy for a client, and you can tell the
00:27:11 --> 00:27:16 AI, hey, please make this professional but casual, detailed but concise,
00:27:16 --> 00:27:20 technical but accessible, comprehensive but focused on key points,
00:27:20 --> 00:27:22 right? And you can see where the problem is.
00:27:22 --> 00:27:27 The AI couldn't possibly satisfy all of these contradictory requirements.
00:27:28 --> 00:27:33 The fix for this problem is to prioritize your most important requirements and
00:27:33 --> 00:27:38 be realistic about what can be accomplished in one output. and also consider
00:27:38 --> 00:27:42 breaking complex requests into a series of simpler prompts.
00:27:43 --> 00:27:47 Mistake number three, forgetting to specify the audience.
00:27:47 --> 00:27:54 Even detailed prompts can fall flat if they don't specify who the content is for.
00:27:54 --> 00:27:58 Different audiences obviously require different approaches, vocabulary,
00:27:59 --> 00:28:05 examples, and tone. The fix here is to always include information about who
00:28:05 --> 00:28:09 will be using, reading, or viewing the final product.
00:28:09 --> 00:28:13 Be specific about their levels of expertise, interest, and needs.
00:28:13 --> 00:28:17 Mistake number four, not providing enough context.
00:28:18 --> 00:28:23 So, AI doesn't know your business, brand, or project history unless you tell it.
00:28:24 --> 00:28:29 If you leave out important context, it will definitely lead to generic results. notes.
00:28:29 --> 00:28:34 The fix for this is to include relevant background information about your brand,
00:28:34 --> 00:28:40 project goals, previous efforts, and any specific circumstances that might affect the output.
00:28:40 --> 00:28:44 Mistake number five, accepting first results without iteration.
00:28:45 --> 00:28:50 One of the big mistakes is definitely treating prompting as a one-and-done process
00:28:50 --> 00:28:52 rather than an iterative one.
00:28:52 --> 00:28:57 Your first prompt rarely produces the best possible result, right?
00:28:57 --> 00:29:00 I caught myself doing this when I'm in a rush to finish a project.
00:29:00 --> 00:29:06 I just accept the first prompt as good enough and I forget that I should definitely
00:29:06 --> 00:29:11 just take a few minutes to refine the prompt or provide feedback on the initial result.
00:29:12 --> 00:29:16 The fix for this is to plan at least one round of refinement.
00:29:16 --> 00:29:21 After getting an initial response, be specific about what's working and what
00:29:21 --> 00:29:25 needs improvement in your follow-up prompt. Mistake number five,
00:29:26 --> 00:29:29 using unclear or ambiguous language.
00:29:30 --> 00:29:34 Words that seem clear to you might be interpreted differently by AI.
00:29:35 --> 00:29:40 Words like modern, professional, or engaging, which we already mentioned,
00:29:41 --> 00:29:44 right, can mean different things in different contexts.
00:29:45 --> 00:29:52 Let's say you're searching for a modern logo concept, and then you're frustrated with the results.
00:29:52 --> 00:29:58 But when you specify modern in the context of minimalist Scandinavian design
00:29:58 --> 00:30:05 with clean lines and subtle color gradients, the output will immediately improve.
00:30:05 --> 00:30:11 So the fix for this is to define subjective terms, provide examples when possible,
00:30:11 --> 00:30:15 and be as concrete as you can about what you mean.
00:30:15 --> 00:30:21 Mistake number seven, not leveraging examples. Many people forget that you should
00:30:21 --> 00:30:25 include examples in your prompts to illustrate what you're looking for.
00:30:25 --> 00:30:28 This is particularly helpful when you're starting a new project.
00:30:28 --> 00:30:35 I remember when I was helping a podcast client develop a format for their show notes descriptions.
00:30:35 --> 00:30:41 Instead of just describing what we wanted, we included an example from another
00:30:41 --> 00:30:46 podcast she liked. And the AI immediately understood and could replicate the
00:30:46 --> 00:30:47 structure with our content.
00:30:47 --> 00:30:54 The fix for this mistake is to include examples of what you're looking for when relevant, right?
00:30:55 --> 00:30:58 Whether that's a style of writing, a type of design concept,
00:30:58 --> 00:31:02 or a particular way of structuring information.
00:31:02 --> 00:31:07 By avoiding these common mistakes, you'll definitely dramatically improve your
00:31:07 --> 00:31:14 results from AI tools. And remember, effective prompting is a skill that develops with practice.
00:31:14 --> 00:31:19 Pay attention to what works and what doesn't and continuously refine your approach.
00:31:20 --> 00:31:26 Let's wrap up this first part of our prompting series by recapping what we've covered today.
00:31:26 --> 00:31:30 So we started by exploring why effective prompting matters so much.
00:31:30 --> 00:31:35 It's literally the difference between getting generic, bland outputs and receiving
00:31:35 --> 00:31:39 useful, creative responses that actually enhance your work.
00:31:39 --> 00:31:44 You can use the same tool and get completely different results based entirely
00:31:44 --> 00:31:46 on how you communicate with it.
00:31:46 --> 00:31:50 We then covered the seven core principles of effective prompting.
00:31:50 --> 00:31:53 1. Be specific and detailed.
00:31:53 --> 00:31:56 2. Provide context and purpose.
00:31:56 --> 00:31:59 3. Specify format and structure.
00:31:59 --> 00:32:06 4. Include examples or references. 5. Use role and perspective prompting. 6.
00:32:07 --> 00:32:09 Iterate and refine. 7.
00:32:10 --> 00:32:15 Use system prompts for consistent interactions. These principles apply across
00:32:15 --> 00:32:17 all AI tools and creative disciplines.
00:32:18 --> 00:32:23 Master these and you'll immediately see better results from any AI assistant you work with.
00:32:23 --> 00:32:28 And then we explored how to adapt these principles for specific creative fields,
00:32:28 --> 00:32:33 whether you're a writer using the CATP framework, context,
00:32:33 --> 00:32:39 audience, tone, purpose, or a designer applying the VCRT approach,
00:32:39 --> 00:32:46 visual context reference technical, or a marketer leveraging the PAISM method,
00:32:47 --> 00:32:49 problem audience insight solution measurement.
00:32:50 --> 00:32:55 Really having structured approaches tailored to your field makes all the difference.
00:32:56 --> 00:33:02 And then we finally mentioned some mistakes you can avoid when prompting the AI tools.
00:33:02 --> 00:33:06 Next week in part two, we'll dive into some advanced techniques that can take
00:33:06 --> 00:33:08 your prompting to the next level.
00:33:08 --> 00:33:12 I'll show you methods like chain of thought prompting for complex problems,
00:33:13 --> 00:33:17 reverse prompting for getting the AI to help you craft better prompts,
00:33:17 --> 00:33:23 and comparative prompting for exploring multiple solutions simultaneously.
00:33:23 --> 00:33:28 Plus, I'll walk you through a detailed demonstration of crafting and refining
00:33:28 --> 00:33:33 prompts for a real project so that you can see the entire process in action.
00:33:33 --> 00:33:37 For this week, I encourage you to practice the fundamentals we covered today.
00:33:37 --> 00:33:43 Pick one of the core principles and really focus on it in your next AI interaction.
00:33:43 --> 00:33:47 Try the discipline-specific framework that applies to your work.
00:33:47 --> 00:33:50 And remember, prompting is a skill that develops with practice,
00:33:51 --> 00:33:53 so don't expect perfection right away.
00:33:53 --> 00:33:58 If you found this episode helpful, please subscribe wherever you get your podcasts
00:33:58 --> 00:34:02 and consider sharing it with another creative professional who might benefit.
00:34:02 --> 00:34:05 You can find show notes with all the frameworks we discussed,
00:34:05 --> 00:34:14 plus example prompts for each creative discipline at www.ai4creativeprose.com.
00:34:14 --> 00:34:17 Thank you for listening to AI for Creative Professionals.
00:34:17 --> 00:34:21 I'm Tiana, and I'll be back next week with part two of our prompting series,
00:34:21 --> 00:34:25 discussing those advanced techniques and real-world techniques.
00:34:25 --> 00:34:47 Music.