Marketing · Email marketing
The Best Prompt for an Email Sequence
The best prompt for an email sequence makes the model plan the arc before it writes. This one asks for a sequence map covering timing, the single job of each email and the objection it answers, then drafts every email with subject line, preview text, body and one call to action. It bans invented discounts, deadlines and testimonials, so the drafts contain only claims you can stand behind, with placeholders where proof is missing.
Email Sequence Prompt
Maps a welcome, nurture, cart or win-back sequence first, then writes each email with one job, one call to action and only the proof you supply.
- Works with
- ChatGPT, Claude, Gemini
- Difficulty
- Intermediate
- Complexity
- Multi-part reply
- Variables
- 7
- Version
- 1.0
Customize this prompt
Fill in what you know. Anything left blank stays as a [bracketed placeholder]. Nothing you type leaves your browser.
7 variables · fields marked * matter most
# Role You are an email marketer who writes lifecycle sequences. You plan the whole sequence before writing a single email. # Inputs - Sequence type: [Sequence type] - Business and product: [Business and product] - Who receives it and what they just did: [Who receives it and what they just did] - The action the sequence should lead to: [Action the sequence should lead to] - Number of emails: [Number of emails] - Facts, offer and proof you may use: [Facts, offer and proof you may use] - Voice: [Voice] # Part 1: Sequence map Before any copy, give a table with one row per email: number, send timing relative to the trigger or the previous email, the single job of this email, the one call to action, and the objection or question it answers. Order the emails by what the reader needs at that moment, not by what we most want to say. Under the table, state the exit rule: what should happen to the remaining emails once the reader takes the goal action. # Part 2: The emails For each email write: - Subject line (50 characters or fewer) and one alternative that takes a different angle - Preview text (under 90 characters) that adds to the subject instead of repeating it - Body: 75 to 200 words, short paragraphs, written to one person - One call to action, phrased as what the reader gets, with the link shown as [LINK] # Rules - One job and one call to action per email. If an email is doing two things, split or cut. - Each email must make sense to someone who did not open the earlier ones. - Use only the facts, offer and proof supplied. Do not invent discounts, deadlines, customer quotes, numbers or guarantees. If an email needs proof we did not provide, insert [PROOF NEEDED: type] and continue. - No false urgency or scarcity. Mention a deadline only if one is in the inputs. - The first email delivers what the reader was promised when they signed up or acted. - Plain language. No "I hope this email finds you well", no "just checking in", no exclamation-mark subject lines. - End every email with [UNSUBSCRIBE LINK] and [SENDER ADDRESS] placeholders. # Output format Part 1 as a table. Part 2 as "Email 1", "Email 2" and so on, each with labeled lines for Subject, Alternative subject, Preview, Body and CTA. Finish with three things worth A/B testing in this sequence and why, described as hypotheses. Send timings are starting points for me to test, so say that once and do not justify them with statistics.
Use it when you are
- Writing a welcome series for new newsletter or lead-magnet subscribers
- Drafting onboarding emails for people who have started a free trial
- Setting up abandoned cart or browse recovery emails for an online store
- Building a re-engagement sequence for subscribers who have stopped opening
Why this prompt works
A map before the copy
Written one at a time, sequence emails tend to repeat the same pitch in different words. The map forces a decision about what each email is for and which objection it handles, and you can reorder or cut rows in seconds before any body copy exists.
One job, one action
Each email is limited to a single purpose and a single call to action, with an instruction to split or cut if it drifts. That constraint is what keeps the emails short and stops the common pattern of a feature tour with four competing links.
Claims are limited to a list you control
Models readily supply a twenty percent discount, a Friday deadline or a glowing customer quote. Restricting claims to the facts you entered, and inserting a [PROOF NEEDED] placeholder where support is missing, means nothing reaches your subscribers that you did not approve.
Every email stands alone
Many recipients will open only some of a sequence. Requiring each email to make sense without the earlier ones prevents openings such as "as I mentioned yesterday" and makes the sequence resilient when timings are changed later. Subscribers who join the story at email three still get a complete message.
An exit rule is part of the plan
Continuing to nudge someone who has already acted is a familiar failure of automated email. Asking for the exit rule alongside the map prompts you to set that condition in your email tool when you build the automation. Without it, the sequence keeps selling to people who have already bought.
How to use it
- Choose the sequence type and describe the trigger precisely: who the reader is and what they did just before the first email.
- Name one goal action. If you have two, run the prompt twice and build two sequences.
- List every fact, offer and piece of proof the emails may use, including real deadlines. Leave out anything you cannot substantiate.
- Check the sequence map first. Reorder or remove rows, then ask the model to rewrite the affected emails.
- Replace all bracketed placeholders, edit for voice, and send test emails to yourself to check subject and preview text on a phone.
- Set the exit rule and timings in your email platform, then test the suggested variations one at a time.
AI model compatibility
| Model | Status | Notes |
|---|---|---|
| ChatGPT | Expected to work | For sequences of six or seven emails, ask for the sequence map first, approve it, then request the emails. This keeps the copy from being cut off and lets you fix the order early. |
| Claude | Expected to work | If you have an email you like the sound of, paste it inside <example> tags under the Voice line and ask the model to match its rhythm, not its content. |
| Gemini | Expected to work | Ask the model to show the character count in brackets after each subject line, then confirm the counts yourself before loading the emails. |
“Expected to work” means the prompt uses plain, model-independent structure and has been reviewed for that — not that it has been run on that model. A model is marked Tested only when a test record is published below.
Worked example
Scenario: An illustrative bookkeeping app that wants trial users to complete setup before the trial ends.
# Role You are an email marketer who writes lifecycle sequences. You plan the whole sequence before writing a single email. # Inputs - Sequence type: Trial or product onboarding - Business and product: Ledgerly, bookkeeping software for freelancers, $15/month after a 14-day free trial - Who receives it and what they just did: Freelance designers and writers who just started the free trial but have not connected a bank account - The action the sequence should lead to: Connect a bank account and send a first invoice within the trial - Number of emails: 5 - Facts, offer and proof you may use: Bank connection takes about two minutes; invoices can be sent from a phone; trial needs no credit card; one real customer quote we have permission to use: "I finally stopped dreading tax season." - Voice: Friendly and direct, like a helpful colleague; no jargon # Part 1: Sequence map Before any copy, give a table with one row per email: number, send timing relative to the trigger or the previous email, the single job of this email, the one call to action, and the objection or question it answers. Order the emails by what the reader needs at that moment, not by what we most want to say. Under the table, state the exit rule: what should happen to the remaining emails once the reader takes the goal action. # Part 2: The emails For each email write: - Subject line (50 characters or fewer) and one alternative that takes a different angle - Preview text (under 90 characters) that adds to the subject instead of repeating it - Body: 75 to 200 words, short paragraphs, written to one person - One call to action, phrased as what the reader gets, with the link shown as [LINK] # Rules - One job and one call to action per email. If an email is doing two things, split or cut. - Each email must make sense to someone who did not open the earlier ones. - Use only the facts, offer and proof supplied. Do not invent discounts, deadlines, customer quotes, numbers or guarantees. If an email needs proof we did not provide, insert [PROOF NEEDED: type] and continue. - No false urgency or scarcity. Mention a deadline only if one is in the inputs. - The first email delivers what the reader was promised when they signed up or acted. - Plain language. No "I hope this email finds you well", no "just checking in", no exclamation-mark subject lines. - End every email with [UNSUBSCRIBE LINK] and [SENDER ADDRESS] placeholders. # Output format Part 1 as a table. Part 2 as "Email 1", "Email 2" and so on, each with labeled lines for Subject, Alternative subject, Preview, Body and CTA. Finish with three things worth A/B testing in this sequence and why, described as hypotheses. Send timings are starting points for me to test, so say that once and do not justify them with statistics.
What a good response should contain
- A five-row map where the first email helps with the bank connection instead of introducing the company
- An exit rule that stops setup reminders once the account is connected and an invoice is sent
- Emails that each carry a single [LINK] call to action and read sensibly on their own
- The supplied customer quote used as written, and [PROOF NEEDED] where an email would benefit from evidence that was not provided
- A trial-ending email that states the real end of the 14-day trial without added scarcity, plus three test hypotheses
This is an illustrative scenario showing how to fill in the prompt and what to look for. It is not a recorded model output. Recorded outputs will be published here once this prompt has a test record.
Testing and editorial review
Status: Untested. Reviewed for structure and clarity by the editorial team. No recorded model test yet.
No test record has been published for this prompt yet. When it is tested, this section will show the model and version, the date, the inputs, the actual output and notes against our seven-point rubric. We do not publish numerical scores.
| Author | TheBestPromptFor Editorial Team — The editorial team writes, reviews and maintains the prompts and guides on TheBestPromptFor. Prompts are drafted with AI assistance and reviewed against published editorial standards. |
|---|---|
| Published | November 15, 2026 |
| Last reviewed | November 15, 2026 |
| Last tested | Not yet tested |
| Version history |
|
Limitations and things to check
- The model cannot see your open, click or conversion data. Timing and order are starting points to test, not findings.
- Email rules on consent, unsubscribe links and sender identification vary by country. The placeholders are a reminder, not legal advice.
- Subject line length that displays varies by email client and device. Send tests before going live.
- Drafts will sound generic until you edit in your own voice, product detail and real examples.
Alternative prompts
Content Writing
Professional Email Prompt
Drafts a single work email that leads with the ask, suits the relationship, and leaves bracketed blanks for any detail you did not provide.
- ChatGPT
- Claude
- Gemini
Content Writing
Conversion Copywriting Prompt
Works out what a page must make the reader believe, then writes the copy in that order using only the proof you supply.
- ChatGPT
- Claude
- Gemini
Marketing
Product Launch Plan Prompt
A two-turn prompt: the model asks its key questions, then builds a launch plan with a message, dated timeline, readiness check, metrics and risks.
- ChatGPT
- Claude
- Gemini
Marketing
Evidence-Based Customer Persona Prompt
Builds personas from the interviews, reviews or survey answers you paste in, tagging every line as evidence, inference or assumption.
- ChatGPT
- Claude
- Gemini
Related resources
Technical guide
Prompt Chaining: Breaking a Task Into Linked Prompts
How to split a large task into a sequence of focused prompts, design the handoffs between them, and debug the chain when a step goes wrong.
Guide
How to Improve AI Responses
A diagnostic approach to weak AI answers: name the fault, trace it to the gap in your prompt, then follow up precisely or rewrite and start again.
Frequently asked questions
How many emails should a sequence have?
Enough to cover the distinct questions a reader has before acting, and no more. Use the sequence map to judge: if two rows have the same job, cut one. Three to five is a reasonable range to start with for most welcome and onboarding flows, then adjust from your own results.
Can I use this for cold outreach?
It is designed for people who have opted in or taken an action with your business. Cold email has different legal requirements and a different job, which is starting a conversation with a stranger. The email writing prompt in the content writing category is a better starting point for one-to-one messages.
What should I put in the proof field if I have no testimonials?
List plain facts: what the product does, how long setup takes, what is included, your refund or cancellation terms. The sequence will lean on usefulness and clarity instead of social proof, and the placeholders will show you where a customer quote would help most once you have one.
Sources and references
- Prompt engineering overview — Anthropic
- Prompt engineering guide — OpenAI
Was this prompt useful?
No account needed. Do not include personal information.