# Generate your first email Creating your first email in Humanic takes less than a minute. You write the intent and Humanic handles the copy, structure, and personalization. **How prompting works** Every email in Humanic starts with a prompt. A prompt is simply you telling the AI what you want — who the email is for, what it should do, and any context that makes it relevant. The quality of your prompt directly shapes the quality of your output. Humanic supports four prompting methods, ranging from quick and simple to detailed and structured. *** ### The four prompting methods **Zero-shot prompting** The fastest way to start. Write two to three lines describing your goal, audience, and context — and Humanic generates a complete email. No examples needed, no configuration required. This is where most people begin and where you should too. **Few-shot prompting** Paste in one or two example emails or sample phrases to show the AI the tone and style you want. Humanic uses these as a reference and mirrors the voice in the output. Best used when your brand has a distinctive style that's hard to describe in words alone. **Chain-of-thought prompting** Walk the AI through your thinking before stating the ask. For example, explain who the customer is, where they are in their journey, and what obstacle you're trying to overcome — then state what the email should do. This method produces the most nuanced, context-aware output and works well for re-engagement or high-stakes lifecycle emails. **Role-based prompting** Give the AI a role before making a request — for example, "act as a lifecycle marketing manager for an early-stage B2B SaaS product targeting developers." This frames every decision the AI makes: tone, structure, calls to action, and messaging priority. Useful when you want consistent output across a series of emails. *** ### Where to start Start with zero-shot prompting. It has the lowest setup cost and produces solid results right away. A strong zero-shot prompt answers three questions: **Goal** — What should this email achieve? Examples: drive feature adoption, re-engage inactive users, announce a new plan. **Audience** — Who is receiving it? Be specific. "New signups who haven't completed onboarding" is more useful than "new users." **Context** — What does the AI need to know? Include relevant product details, timing, tone preferences, or campaign background. You can write this freeform or choose one of Humanic's preset prompts to get started immediately. Once your prompt is ready, click Enter and Humanic will generate the email in seconds. *** ### Refining your output Your first generation is a starting point, not a final draft. If the output isn't quite right, don't start over. Just reply in the same conversation with additional context or a correction — Humanic refines based on the thread. Most users get to a publish-ready email in two to three exchanges. Prompt-based email generation is a method of creating email content by describing your goal in\ plain English rather than filling in a template. You type a short instruction — the prompt — and\ the AI produces a complete email: subject line, preview text, and body copy.\ \ **How it differs from traditional email builders?**\ \ Traditional email platforms (Mailchimp, Klaviyo, Brevo) require you to select a template, drag\ and drop blocks, manually write copy, and configure segments separately. Prompt-based\ generation collapses all of these into a single step: describe what you want, and the platform\ produces a ready-to-send email.
1. **Zero Shot**: A single instruction describing your goal. Best for simple campaigns. 2. **Few Shot**: Include 1–2 example emails in your prompt to guide tone and format. 3. **Chain of Thought**: Break your prompt into steps ("First explain the offer, then add urgency,\ then include the CTA"). Best for complex or multi-section emails. 4. **Role-Based**: Assign a persona to the AI ("You are an email marketer for a DTC skincare brand..."). Best for maintaining consistent brand voice. It usually takes about 8–10 prompts to create your first email. Start with a simple prompt, then take a few minutes to refine your goal and decide if the email is ready to send to your users. Typically the first email will be very good but it might not have everything correct, you can use the chat window to continue to iterate and prompt to make it exactly what you want. Note: Humanic does not have a drag and drop editor but you are able to view the MJML code to edit if the models are mis-behaving. We host weekly Learn to Prompt to Write Email Campaigns That Convert. Click here to learn more and register for an upcoming session --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://humanic.gitbook.io/humanic/getting-started-with-humanic/generate-your-first-email.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.