The Launch Week Email Kit
Five emails, one deadline, and the Thursday email that tells the wrong buyer to keep their money.
Last week you built the offer. This week you sell it.
Drop 08 left you with offer-stack.md: your core outcome, the bonuses aimed at real obstacles, a guarantee your margin can survive, and urgency that is true. That file is finished and it is sitting on your computer. Nobody on your list has heard about any of it.
The Launch Week Email Kit is six prompts and one file called launch-week.md. Prompt 1 builds the launch brief and does one thing first: it tests your deadline. It classifies what you have as COHORT, PRICE CHANGE, BONUS EXPIRY, CAPACITY, or SEASONAL, and if it fits none of those, it stops and tells you that you do not have a launch, you have a week. Prompt 2 drafts the week straight out of your offer stack and your proof bank, with word count targets per email, one link per email, and a hard rule that any number it cannot trace to your proof comes back as [NEED NUMBER] instead of an invention. It leaves Email 4 alone, because Email 4 has its own prompt and its own rules. Prompt 3 gives you eight subject lines and three preview text options per email, ranked, with a flag on any line that would read as a lie the second the reader opened it. Prompt 4 writes the Honest No. Prompt 5 is a separate proofreading pass over the finished set: claims against proof, the deadline identical in all five, links, repetition, the read aloud test, weekday references, and spam risk. Prompt 6 runs the Monday after close and tells you which email to rewrite from scratch next time.
The send calendar is in the file. Monday through Friday, two sends on the last day, and the segmenting rule for people whose email tool cannot segment.
The email most people skip
Thursday, halfway between the proof and the close, you send under 200 words telling a specific reader not to buy.
Not the coy version. A real one, naming who this does not fit, what it costs them in their own hours, and where to go instead when their problem is a different problem. By Thursday every message in your reader’s inbox is arguing for the sale, including yours. An email that argues against it, and gives up a sale on purpose, is the only evidence available that you are describing your product accurately.
It fails when it is fake. “Not for people who aren’t ready to succeed” is not a disqualifier, it is a dare, and the reader can tell which one they are holding. The prompt has a hard rule for exactly this: every disqualifier has to cost you a real sale, and any line that disqualifies nobody gets cut.
Why the manual way costs you
Launch emails are the piece people hand to someone else, and the invoice lands before anyone writes a sentence. A freelance copywriter charges you $1,000 to $5,000 for a launch sequence in typical 2026 US rates, or $100 to $500 an email when you buy them one at a time, so the six sends in this calendar run you $600 to $3,000 that way. That copywriter still needs your proof, your objections, and your real deadline before they can start. You pay the invoice and you still supply the hard part.
The software takes its cut on the other side. Kit’s Creator plan charges you $39 a month at 1,000 subscribers, $33 a month if you pay for the year, so $396 a year out of your pocket to send six emails to people who already subscribed to hear from you. Kit’s free plan sends broadcasts to as many as 10,000 subscribers, and a six-broadcast launch week does not need the automation the free plan limits. Both numbers were checked today.
Running all six prompts end to end moves about 50,000 input tokens and 25,000 output tokens, which at Claude Sonnet 5 pricing of $2 per million input and $10 per million output, verified today, is 10 cents in and 25 cents out. Thirty-five cents for the week. Inside a Claude subscription you already pay for, it is $0 in new tools. Thirty-five cents against a $1,000 sequence, and the file you keep makes your second launch a 40-minute edit of your first.
The cost with no invoice is the bigger one. Right now your offer exists and your list does not know. Every week that stays true, you are paying to reach strangers on Meta while the people who already raised their hand hear nothing from you about the one thing you built for them.
How to use it
- Run Prompt 1 today, before you write anything. Twenty minutes. If it tells you your deadline is not real, believe it and make the smallest business change that creates one. A launch week without a close is five emails asking people to think about it.
- Run Prompt 2 with
offer-stack.mdandproof-bank.mdpasted in. You get five drafts that already agree with your sales page, because they came from the same file it did. Fix the [NEED NUMBER] flags by going and getting the numbers, not by softening the sentences. - Run Prompt 4 on its own and send what it gives you. If you read the Honest No and feel a small flinch because one line describes people who have paid you before, that is the email working.
- Run Prompt 5 the day before you send, in a fresh chat. Then put Prompt 6 on your calendar for the Monday after close, on the same day you update your proof bank with whatever this launch produced.
Eight drops built the thing you sell and the page that sells it. This one points all of it at the people who already asked to hear from you, which is the cheapest audience you will ever have and the one most small businesses never actually pitch. Your offer has been finished since last Friday. Friday night it either sold something or it did not, and you will know which. Either it makes you money or it costs you nothing.
Next Monday’s Tool Teardown: verdict as always, one real test, keep or replace.
Next Friday’s drop: the Post Purchase Sequence. The seven emails between “thanks for your order” and a review request that actually gets answered, built out of your own delivery steps.