ChatGPT Email Generator

The garbage-in-garbage-out problem

ChatGPT and other large language models are extremely capable email writers — but only when they have enough context to know what kind of email to write. Asking ChatGPT to "write a follow-up email" produces a generic, fill-in-the-blank template because that's the safest output in the absence of context. Asking it to "write a follow-up email from the head of growth at a SaaS company to a VP Marketing they met at SaaStr last month, offering a 15-minute intro call" produces a usable draft on the first try. This tool wraps the second pattern into a form so you don't have to remember the structure each time.

Specifying the goal explicitly

Eight common email goals — cold outreach, sales follow-up, customer support, newsletter, internal update, rejection, thank-you, announcement — cover the vast majority of professional email writing. Each goal has a different shape: cold outreach needs a specific observation about the recipient and a low-friction ask; rejections need to be direct and respectful; thank-you notes need one concrete specific detail. The prompt builder encodes the structural guidance for each goal into the prompt itself, so ChatGPT produces the right shape on the first generation rather than the fifth.

Tone as a first-class input

Tone is the single hardest thing for an LLM to get right without explicit guidance. The default ChatGPT email leans corporate-friendly: polite, slightly formal, no personality. For most professional email this is fine; for cold outreach, support replies, or internal team email it's flat. The six tones in this tool — professional, friendly, casual, formal, persuasive, concise — each include specific stylistic guidance (no exclamation marks, contractions are fine, strip every line that doesn't earn its place) that ChatGPT respects when it's spelled out explicitly in the prompt.

Length enforcement

ChatGPT defaults to medium-length emails (180–250 words) almost regardless of context. That's too long for most cold outreach, too long for most follow-ups, and sometimes too short for newsletters or detailed support replies. The three length presets in this tool — short (60–90 words), medium (120–180 words), long (200–300 words) — enforce explicit word-count ranges in the prompt, and ChatGPT respects them reliably. Choose short for follow-ups and quick asks; medium for most things; long when the email genuinely needs to do explaining work.

Constraints — the secret weapon

The constraints field is where the prompt gets sharpest. "No emoji" produces a noticeably different email from the default. "Include a placeholder for a calendar link" ensures the CTA is structurally sound. "Don't reference our previous email" saves you from awkward self-references when sending into a new thread. The constraints field is optional but worth using whenever you have a specific preference — even one short line dramatically improves the output's alignment with what you actually want.

Bring your own model

The prompt this tool generates works on every modern LLM: ChatGPT, Claude, Gemini, Llama, Mistral. The structure is model-agnostic. The "Open in ChatGPT" button is the convenient default because ChatGPT has the largest user base, but you can paste the same prompt into any chat interface and get equivalent results. This is intentional — locking the prompt to a single model would defeat the purpose of an open tool.