How to Write a Cold Email Sequence (Annotated Example)

Anyone can copy a cold email. Paste it, swap the company name, hit send. What they can’t copy is the reasoning that put every line where it is, and that reasoning is the whole game. A sequence you copy works until your offer changes, then it falls apart, because you never knew which parts were load-bearing. A sequence you understand travels to any offer, any market, any list.

So this is a teardown. One offer, five emails, and the why behind every decision.

A note on the offer: it’s invented. We’re an open book about how we work and a closed book about who we work for, so we never publish a client’s campaign. The pitch below, recovering failed subscription payments for DTC brands, is made up. The method is exactly what we run.

The shape

Each prospect gets three touches on the same subject line, so the whole thing lands in one thread and reads like a person following up, not a system firing: an opener, a follow-up two to three days later, and a handoff a few days after that.

The opener is where we test. We write two or three versions of that first email, each built on a different principle, and let the replies tell us which angle the market wants. That is the five emails here: three opener variations, one follow-up, one handoff.

The subject line: lowercase, one to three words

The subject is two words, lowercase, no pitch: failed payments. A subject line is a thumbnail, not a headline. Its only job is to look like an internal note a colleague would send, because that is the email people open.

A promo-looking subject like “Recover 9% of your lost MRR today!” gets grouped with the 40 other promos in the inbox. Something that reads like a coworker flagged it gets opened.

And we reuse the exact same subject on every email. That keeps the sequence in one thread, so by the third touch the prospect has a growing conversation with one subject line, not three disconnected cold emails. Threading builds familiarity, and familiarity gets replies.

The opener, three ways

We never send just one version of the first email. Each variation tests a single idea. Here are all three.

Variation A: open with a question

Good morning {{first_name}},

Quick one: how much does {{company}} write off each month to failed subscription payments?

For most DTC brands we see, it’s 5 to 9% of recurring revenue, mostly expired cards that never get retried at the right time.

We recover a chunk of that automatically, with no dev work on your side.

Worth a quick look?

Lowercase subject, opening question. A question earns a reply better than a claim does. It hands the prospect something to answer instead of something to evaluate, and it makes them look at a number they probably have not measured, which is exactly the gap the offer fills.

Variation B: personalized first line

Hi {{first_name}},

Saw {{company}} runs on a subscription model, so this should land.

Brands your size quietly lose 5 to 9% of recurring revenue to failed charges every month. That’s real money on the table, not a tech project you need to rebuild for.

We plug the leak in about a week.

Open to seeing the numbers for {{company}}?

The personalized first line is the hook. It is the only line that has to feel written for this one person. Everything after it can be shared across the list. This is where personalization actually pays: one sentence that proves you looked, not a whole email of forced detail.

Lead with the outcome, not the feature. It opens on the revenue you are losing and treats the mechanism, how we recover it, as a footnote. Nobody buys automatic card-retry logic. They buy the money it brings back.

Variation C: one idea, one CTA

Hey {{first_name}},

Failed payments are probably costing {{company}} more than your whole retention budget saves.

We fix that without adding work for your team.

Reply “send it” and I’ll share a two-minute breakdown of what recovery could look like for you.

Under 100 words, one idea, one ask. The call to action is deliberately low friction: “send it” costs the prospect nothing and commits them to nothing. We are not asking for a call yet. We are asking for permission to be useful, and the call comes after the value does.

Across all three, one rule holds: pick one focus per email, outcome or mechanism, never both. The moment an email argues two things, it argues neither.

The follow-up: add proof, not a new pitch

Two to three days later, same thread:

Hi {{first_name}},

Following up on the note about recovered revenue.

One brand we work with was losing about $40k a month to failed charges. We got roughly 60% of it back in the first 30 days, with no engineering lift on their end.

The same playbook would apply to {{company}}.

Want the short version?

P.S. happy to send the breakdown even if the timing is off, no strings.

One proof point, not a new pitch. If the first angle did not land, a second angle just reads as noise. So this email does one thing: it adds a single proof point. One brand, one number, one timeframe. Specific and checkable beats broad and impressive.

The P.S. is the second hook. Plenty of people skim to the P.S. first, so it earns its own soft offer: the breakdown, no strings. It gives a reason to reply even to someone who is not ready to buy.

The handoff: route, don’t re-sell

The last email is the shortest, and it does not pitch at all:

{{first_name}}, last note from me on this.

Are you the right person to talk recovered revenue at {{company}}? If not, no worries, just point me to whoever owns retention.

If it’s not a priority right now, I’ll close the loop on my end.

Route to the right person, don’t re-pitch. The routing question often gets forwarded internally, which is a warmer intro than any cold email. And the low-pressure exit does something counterintuitive: giving someone an easy out is what makes them reply. Pressure gets ignored. Permission to say no gets answered. If they wanted the pitch, they had it already. This email is about resolution.

Cadence and testing

The three touches run on a schedule: the opener on day zero, the follow-up two to three days later, the handoff three to four days after that. The delays are relative, so the follow-up waits two days from the opener, not two days from launch. Short enough to stay top of mind, long enough to not nag.

We never ship one version. Every step gets two variations, and each variation tests exactly one idea, never five changes at once. On the opener that means the three angles above, question versus personalized versus one-idea. Elsewhere it is one lever at a time: a time-boxed call to action against an open one, with a lead magnet against without, a pain-led hook against an outcome-led one.

Greetings, transitions, and sign-offs run through spintax, so a line like {Hey|Hi|Hello} {{first_name}} varies naturally across sends and no two inboxes get an identical footprint. Merge tags like {{first_name}} and {{company}} carry the personalization, and every variation is proofed against real sample data before a single email goes out.

Adapt this to your offer

Strip out failed payments and DTC brands, and here is the skeleton that remains. Drop your own offer into it and the sequence holds.

Subject. One to three lowercase words that sound like an internal note. Reuse the exact same one across all three touches so they thread.

Opener, day 0. One focus only, pick pain, outcome, a question, or proof. Under 100 words. One low-friction call to action.

Follow-up, day 2 to 3. One proof point, one number, one timeframe. No new pitch. Add a P.S. with a soft, no-strings offer.

Handoff, day 5 to 7. The shortest email. Ask if they are the right person, or who is, and give a clean, low-pressure way out.

Everywhere. Lead with the outcome, keep the mechanism underneath, and change one variable at a time when you test.

The principle underneath

Every choice above ladders up to one idea: the reasoning is the asset, not the copy.

Look at what each rule protects. Lowercase subject, question opener, outcome over mechanism, one idea per email, proof over pitch, route over re-sell. None of those are specific to failed payments or to DTC brands. Swap in a different offer and every principle still holds, even though every sentence changes. That is the test of a real method: the words are disposable, the reasoning ports.

This is why we publish the sequence with the why written in. A competitor can copy the copy. They cannot copy the judgment that decides, on a new offer with a new list, which line is load-bearing and which is filler. That judgment is what you are hiring when you hire us, and it is the part that compounds: every campaign we run sharpens it.

The email is the thing you see. The thinking is the thing that works.