How to Format Data for AI-Generated Emails

A few weeks ago, I asked ChatGPT to help me write a professional bio. What did it give me? “Hello. My name is Edward. I enjoy work. Also, computers.” Utter nonsense. Not because the AI is broken. But because I asked it like I was ordering a pint in a power cut. Yes, AI is clever. But more importantly? It's obedient. Usually. You get rubbish in → you get rubbish out. So if you want useful results, you need to master the underrated art of Prompt Engineering. It’s not technical. It’s just how to talk to robots so they don’t embarrass you. Here’s your no fluff starter pack: 1) Be specific ↳ “Write something” = confusion ↳ “Write a 3-line cold email to a CFO who hates fluff” = clarity 2) Give context ↳ AI needs background, not guesswork ↳ Who’s it for? What’s the goal? What do you do? 3) Show examples ↳ “Make it like this” is miles better than “Make it good” ↳ Even machines need mentors 4) Set the format ↳ Want bullet points? A snappy tweet? A poem? ↳ Say it. Don’t assume. 5) Use frameworks ↳ Try: Role + Task + Context + Format + Tone ↳ “You’re a hiring manager. Write a LinkedIn post for interns. Make it short, warm, and slightly smug.” 6) Define the tone ↳ Friendly? Formal? Passive-aggressive manager email? ↳ Pick it. Name it. 7) Ask follow-ups ↳ First drafts are just that: first ↳ Don’t settle. Steer it. 8) Tweak mercilessly ↳ Change one word. Watch it change everything. ↳ Prompting is editing in disguise. Bottom line: If AI is giving you rubbish, it’s not broken. It’s just following orders. So give better ones. 📣 Tag someone still typing “write me content” like it’s 1997. And if you're brave enough… Challenge them to a prompt-off. Loser writes their own emails this week. Follow Edward Frank Morris for more posts like these.

Anatomy of a Prompt Want better results from AI? It starts with building the right prompt. Shocking, right? I've been learning lately how to work better with Prompt Builder. Here's the breakdown: 1. Instruction: The foundation of every good prompt. ↳ Definition: The core task or goal you want the AI to perform. ↳ Purpose: Ensures the AI knows exactly what to do. Example: "Write a formal email to follow up with a customer about a recent purchase." When your instructions are clear, your results are too. 2. Locale, Style, Tone: Speak their language. ↳ Definition: Adjust the geographic, cultural, stylistic, or emotional elements of your output. ↳ Purpose: Tailors the response to your audience. Example: Locale: "Use British English spelling. Write in a conversational style. Be friendly and supportive." Personalized communication creates a connection. 3. Hyperparameters: Fine-tune for precision or creativity. ↳ Definition: Settings like creativity level, response length, and repetition penalty. ↳ Purpose: Customize how the AI behaves. Example: Creativity: "Set temperature to 0.8 for more creativity." Length: "Limit the response to 150 words." 4. Policies: Stay within the lines. ↳ Definition: Guidelines or restrictions for your AI output. ↳ Purpose: Ensure compliance and professionalism. Example: "Do not include personal information." "Avoid using jargon unless specified." Set boundaries to build trust. 5. Output Format: Structure matters. ↳ Definition: Define how the AI presents the response. ↳ Purpose: Improve clarity and readability. Example: "Provide a numbered list of steps." "Summarize the details in a short paragraph." Clean formats = faster action. 6. Model Choice: ↳ Definition: Pick the AI model that fits your task. ↳ Purpose: Leverage the unique strengths of different models. Example: ChatGPT o1 7. Examples: Show, don’t just tell. ↳ Definition: Provide sample inputs and outputs, negative and positive ↳ Purpose: Clarify expectations and improve quality. Example: "Here's a good example for an email response: [...]" "Here's a bad response: [...]" Examples helps align the output to your liking. 8. Interaction Context: Keep it relevant. ↳ Definition: Give the AI context about the task or conversation. ↳ Purpose: Maintain continuity in responses. Example: "The user is asking follow-up questions about their order placed last week." Context creates seamless, personalized interactions. 9. Grounded Data: Anchor outputs in facts. ↳ Definition: Use specific, factual data from your database to guide responses. ↳ Purpose: Enhance reliability and personalization. Example: "Pull customer name and order details from Salesforce records." "Reference statistics from a trusted source." Grounded data = reliable, actionable insights. What have you found makes the most impact on your prompts, whether using Salesforce or not?

When crafting AI prompts for outbound emails, include examples of Input & Output pairings to dramatically improve your results 🚀 Why? AI looks for patterns in the data you feed it 🤖 . When you provide examples, you also provide it with a pattern to follow. Without them, it can easily misunderstand what you're asking it to do. Let's look at an example: You're reaching out to 50 companies with sustainability targets in their ESG reports. To capture their attention, you want to insert a hyper-personalized first line that references these targets. In your prompt, provide instructions on the writing task. Then include a section called Examples at the bottom. It should look something like this: Examples: **Input**: Acme’s COO, Valerie Williams, announced several sustainability targets. One included being carbon neutral by 2030. **Output**: I was just reading your recent ESG report and noticed Valerie’s announcement of Acme aiming to become carbon neutral by 2030. Providing 3 input & output pairings will see you consistently getting high quality outputs that you can insert into your outbound messaging 📈 Need help refining your prompts? Send them my way, happy to review!