Generate on-brand email copy with AI

SaaStr made one thing painfully clear: Everyone is using AI. But only a few have a real process. That’s why one session stood out. Not only for the best slides, but for the content on them. It was from my alma mater, Navneet Singh, now VP of Marketing at Palo Alto Networks. His talk was on AI in marketing  and it didn’t repeat the usual stuff. No “AI will change everything” slides. No ChatGPT worship. Instead, he shared the exact framework they use internally to write better prompts, faster: RISEN. Here’s the breakdown: R – Role - Tell the AI who it is (copywriter, marketer, strategist) before it begins. I – Input - Feed it real context. Data, audience insight, actual messaging problems. S – Steps - Break down what you want in steps. Don’t just say “write a blog.” Tell it how. E – Examples - Show the tone, format, or structure you’re aiming for. AI imitates patterns better than it invents them. N – Nuance - Tell it what to avoid, what not to say, or how to sound, this is where prompts go from generic to gold. Navneet also shared two tactics that stuck with me: 1. Chaining multiple AI tools Instead of using one tool for everything, they combine several: → One for idea generation → One for tone refinement → One for fact-checking or pulling internal data Faster, cleaner, more reliable output. 2. Training custom GPTs to scale brand voice Fine-tuned on internal docs and past campaigns, so anyone across the org can generate content that sounds on-brand - instantly. We’ve already started testing this at Mailmodo. Big takeaway: AI doesn’t need hype anymore. It needs systems.

5 ways I use ChatGPT to actually make my copywriting better: 1. Gather snippets of copy from the brand's website, social media and other sources. Like captions, about us pages and more. Then I ask AI to analyze the voice and tone. And describe it in 5 words. Then I use the same 5 words to guide my copy if I'm going off-brand. 2. Upload reviews and testimonials in a CSV file and ask GPT to analyze them. Then I ask it to pull out the common sentiment, emotional triggers, most pressing problems, stand-out benefits and how people have worded their experience. This helps me discover emotive language for headlines. 3. If I'm writing for the same brand, I give chat 1-3 previous examples to take inspiration from. This works well for email campaigns. Always put in this instruction at the end: Don't copy word for word and just take inspiration from the email in terms of style, format and length. 4. I love playing the constraints game with ChatGPT. I never ask it to just write copy. I focus on: Word count/length A framework (like PAS) Themes (lead with emotion vs lead with benefit/USP) My creativity thrives when I give it constraints because I can see how the same thing can be communicated in different ways. 5. Use it as a first draft generator I never let it write the final copy but I do let it help me beat writer’s block. If I’m stuck on a hook, I’ll give it a clear prompt with the target audience, product benefits, theme and tone of voice. Then, I ask for 10 variations. Out of those 10, maybe one or two are solid. The rest? Either too generic or off-brand. But that’s the trick. ChatGPT isn’t here to replace my creativity. It’s here to spark it. It helps me go from 0→1 faster. And considering I write for multiple brands in the same day, the extra brainstorming power helps. How do you use it? Let me know in the comments!👇🏻 Follow #OKCreative for more. #copywriting #writing #ChatGPT #ai

Early-stage startups struggle to keep their brand voice consistent when they scale content. Here's the exact prompt engineering framework that solves it 👇 You hire writers. Brief agencies. Try templates. But nothing sounds quite right. Here's what most founders miss: You don't need more people. You need a better prompt. I spent weeks training Claude to write in my exact brand voice. The result? Content that converts like our best sales pages, without me writing every word. Here's the exact prompt engineering framework I use for myself & my clients: Step 1: Define Your Baseline Context Start every prompt with role clarity. Example: "You are a CMO for tech startups writing to technical founding teams." This isn't optional. It sets the entire frame for how the AI approaches your content. Step 2: Upload Your Voice DNA Don't just describe your voice, show it. What to include: → 6 brand voice adjectives (approachable, practical, conversational, clear, encouraging, expert-yet-humble) → Full example posts that converted (at least 20,000 words) → Writing rules (like "never use em dashes" or "never write 'It's not X, it's Y'") The AI needs to see patterns, not just hear descriptions. Step 3: Layer In Strategic Context This is where most people stop too early. Don't just give Claude voice, give it strategy: → Your differentiation (what makes you different) → Your frameworks (methodologies you use) → Your positioning (how you want to be seen) → Case studies and social proof Now the AI isn't just writing in your voice. It's writing from your strategic position. Step 4: Build Guardrails Tell the AI what NOT to do. Example rules: → Never talk down to people → Speak in a supportive voice → Avoid corporate jargon → Don't use clichés or buzzwords These constraints keep output on-brand even when the topic shifts. Step 5: Test and Iterate Run the prompt. Compare output to your best-performing content. Adjust the framework based on what's missing. Iterate 10-20 times until it sounds like you. What this actually looks like in practice: Week 1: Output sounds generic and robotic Week 2: Getting closer, but missing your edge Week 3: Starting to sound like you Week 4: Content you'd proudly publish under your name If your early-stage startup is drowning in content creation, this framework changes everything. ♻️ Repost this to help founders scale their brand voice. 🔔 Follow me for more AI-powered content strategies. P.S. What's the hardest part of your brand voice to replicate?

I've spent 12 months building AI email marketing systems at Impact Theory. Here's how we reduced our marketing team head count by 75%, with 3x the results. What did we learn? We cut our marketing team from 4 people to 1. We increased output by 300% with better quality. We turned email marketing into a predictable system that can almost run without me. Most people get AI copywriting completely wrong. They ask for "good copy" and wonder why it sounds generic. They one-shot prompts without any context. They treat AI like a magic content machine instead of a junior writer who needs direction. Those who are getting insane results with AI copy take a different approach. I use what I call the "Voice Cloning System": Step 1: The Voice Training Upload 100+ hours of coaching call transcripts. This teaches AI your exact frameworks, language patterns, and how you explain complex ideas. Step 2: The Specialist GPTs Build separate GPTs for weekly newsletters, bite-sized value emails, and email sequences. Each one knows its specific job and audience. Step 3: The Framework Extraction Prompt: "Analyze these coaching transcripts. Extract my top 10 frameworks and how I typically explain them. Turn these into high-value frameworks." Step 4: The First Draft Generator Prompt: "Write a weekly newsletter about [topic] using my voice and the [framework from step 3]. Keep it conversational but authoritative." Step 5: The Human Polish Don’t skip this! AI gives you the 80% draft. You finesse the final 20% that makes it unmistakably yours and converts like crazy. The nuclear question: "What would make this email sound exactly like me teaching this concept on a coaching call?" This system saves us 20 hours a week and creates emails with a higher ROI than our old team ever achieved. In a world where everyone's emails sound like AI wrote them, the advantage goes to those who train AI to sound exactly like them. Most people use AI as a replacement writer. Be the one who uses it as a junior copywriter who knows your voice better than most humans. AI just changed what’s possible. See if your idea is ready to launch in minutes with my Zero to Launch GPT: https://buff.ly/lYxdOHG

ChatGPT writes terrible marketing emails. Unless you know how to prompt it. Here's the 3-step system I use for every campaign: Most marketers type "write me an email" and get generic garbage. That's because AI doesn't know your brand voice. You have to teach it first. 𝟭. 𝗧𝗿𝗮𝗶𝗻 𝗶𝘁 𝗼𝗻 𝘆𝗼𝘂𝗿 𝘃𝗼𝗶𝗰𝗲: Paste at least 3-5 emails that performed well (mix of welcome, promo, and post-purchase if you have them). Pro tip: Don't pick your cleverest emails. Pick the ones that converted. AI needs to learn what works, not what sounds good. Then use this exact prompt: "Here are my best-performing marketing emails. Analyze the tone, sentence structure, word choices, and formatting patterns. This is my brand voice. Use it as the foundation for every email you write for me. Confirm you understand before I give you the next task." Now it has a reference point. 𝟮. 𝗨𝘀𝗲 𝘁𝗵𝗶𝘀 𝗽𝗿𝗼𝗺𝗽𝘁 𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲: "Write a [type] email for [audience] about [topic]. Goal: [click / purchase / reply]. Tone: [casual / direct / warm / urgent]. Length: [short / medium]. Use my brand voice from above." Specific inputs = less editing later. 𝟯. 𝗧𝗵𝗲 𝗵𝘂𝗺𝗮𝗻 𝗲𝗱𝗶𝘁𝗶𝗻𝗴 𝗽𝗮𝘀𝘀: AI gets you most of the way there. The last pass is you. Here's what generic AI output looks like vs. edited: ❌ "We're excited to announce our new collection!" ✅ "The summer drop is here." ❌ "Don't miss out on this limited-time opportunity!" ✅ "Ends Friday." ❌ "Click here to learn more about our products." ✅ "See the collection." Swap generic AI phrases for how your brand actually talks. Read it out loud. If you wouldn't say it, cut it. That's where the robot disappears. Why this matters: Generic AI emails don't just sound bad. They train your subscribers to ignore you. AI isn't a replacement for good email marketing. It's a first draft machine. You bring the strategy. The voice. The thing that makes subscribers actually buy. Save this for your next campaign. ♻️ Repost to help a fellow marketer. Follow me, Chase Dimond, for more email marketing + AI systems.