Cold Email Sequences Written Around a Single Claim

The number a rewrite moves is positive reply rate. Between one and three percent of contacted people replying with interest is a working B2B sequence. Below half a percent the fault is almost always the claim rather than the sentences. We do not report open rate: Apple's Mail Privacy Protection has been firing tracking pixels automatically since September 2021, and we recommend turning the pixel off entirely, which removes the number anyway. One claim, four times: A sequence is one argument delivered four times, not four unrelated ideas hoping one lands. We write the claim first, in one sentence, and if your team cannot agree on that sentence we stop and work on it, because no amount of rewriting fixes a sequence that is arguing with itself between touch two and touch three. The shape we write to: Touch one, day zero. Sixty to ninety words. No link, no image, no attachment. It ends on a question answerable in one line. It states the claim and cites the evidence field from your list, the tender award or the patch note or the hiring signal, so the recipient can see why they specifically were written to. Touch two, day four, same thread. The proof. One number from a comparable engagement or a public source, named and datable. Nothing we cannot source goes in, including your own case study figures if the client will not let us name them. Touch three, day eight, new thread and new subject. The same claim from a different buying reason. Cost, risk and headcount are three different arguments aimed at three different people inside the same company. Touch four, day twelve. Two lines, permission to close the file, no manufactured guilt and no false deadline. Subject lines: Two to four words, sentence case, no brackets, no emoji, no fabricated RE: or FW:. Anything that reads as a trick buys an open and loses the reply, and at your price point the open was never the asset. Replies are part of the copy: You get six replies and only six. Not interested, send information, wrong person, bad timing, we already use a competitor, and what does it cost. We write all six. The one that pays is wrong person: a referral request written properly converts several times better than the cold approach that produced it, and most teams answer it with a thank you and lose the account. How we test: One variable at a time, and no result is read below 400 contacted per arm. We test the claim before the subject line, because the claim moves the number and the subject line is what teams change when they have run out of ideas. We will not declare a winner on eighty sends, and we will push back when your dashboard says we should. Language: Outbound to Indian recipients runs in English. We write and speak Hindi and use it where it belongs, in call follow-up and WhatsApp. A Hindi cold email to a procurement head in Gurugram reads as a mass send, and we would rather say that than sell it back to you as localisation. What we refuse: Spintax. Merge tags beyond first name and company. Openers built on someone's LinkedIn posts by a model, which every buyer at this level has now seen several hundred times. Breakup emails implying the recipient owes you a reply. Claims about your product we cannot trace to something you have shipped. What you receive: The sequence, a variable map matched column by column to your list, the six reply scripts, and a one-page rules sheet stating what a rep may change and what they may not. Who this is not for: Teams who have not decided what they sell. Anyone expecting copy to rescue a list built from filters. Pre-launch products with nothing shipped to point at.