How to Write a Research-Based Cold Email (With a Real Structure)

A research-based cold email has one job before it has any other: prove, in the first line, that the sender actually looked at this specific company. Everything else, the pitch, the offer, the ask, only gets read if that first line earns it.

The structure

1. Open with the fact, not a greeting-plus-preamble.State the real, current thing you found, plainly. No “I came across your profile and…”, no warm-up sentence before it.
2. Explain what you do, in two sentences. Plain language a reader can understand on a phone, between meetings. Not a feature list.
3. Lower the risk of saying yes. Whatever makes trying it easy: a free pilot, a short call, a sample rather than a commitment.
4. Ask one thing.A single, specific question. Two asks read as indecisive, and force the reader to choose, which they usually won’t bother doing.

A worked example

The company, person, and detail below are invented for illustration.

Hi Sam,

Saw Northfield Robotics just brought on your first Head of Growth, three weeks after shipping the new integrations tier. That’s usually the exact point outbound either gets built properly or falls off the plate entirely.

We do the research and drafting work that normally eats the first few months of a growth hire’s time: finding the right prospects, verifying their contact details, and writing the first email from something real about their company.

Happy to run it free on your first list, no commitment either way. Worth 15 minutes this week?

Why length matters

Short isn’t a style preference, it’s a filter. A long email asks for more of the reader’s time before they’ve decided whether to give any. Aim to say what you need in well under 150 words, and treat anything longer as a sign there’s a sentence in there doing no work.

Common mistakes that break the structure

• Burying the hook.If the researched fact isn’t in the first two lines, most readers won’t get to it.
• Explaining the product before proving relevance. Nobody reads a pitch from a stranger before deciding whether the stranger has actually looked at them.
• Two calls to action.“Worth a call, or happy to send more info” gives the reader an easy way to do neither.
• A stale hook. See which signals are actually worth using, and how to judge whether one’s still fresh.

Where the research comes from

None of this works without real research behind it. See how to research B2B prospects before you write anything for how to find the fact this whole structure depends on.

Why ninety words, and not a hundred and fifty

Short is not a style preference, it is a filter. A cold email is a request for attention from someone who owes you none, and every additional sentence is another opportunity to be closed. Ninety words is a working ceiling for a first touch; if you cannot make the case in ninety, the problem is usually that you are making two cases.

A practical way to find the fat: read the draft and mark every sentence that would still be true if you sent it to a competitor. Those sentences are doing no work for this recipient. Most first drafts lose a third of their length to that one pass.

Subject lines

Match the subject to the observation in the body, keep it to three to five words, and write it in lower case as you would to a colleague. “your revops hire” is a better subject line than “Transforming Your Outbound Strategy” for the same reason the opening line works: it reads as though a person had a specific reason to write.

Avoid curiosity-gap subjects (“quick question”, “re: our conversation”, “following up” on a thread that never existed). They get opened and then resented, and resentment shows up in replies rather than in opens, which is one reason open rate is the least useful metric in cold email.

The ask, and why it should cost less than the email did

The most common way a well-researched email fails at the last line is by asking for thirty minutes. You have spent ninety words earning perhaps four seconds of goodwill, and then requested half an hour of diary time from a stranger. The arithmetic does not work.

Ask something that can be answered in one line instead. “Is this something you own?” “Worth a look, or is this handled?” A reply of any kind is the actual objective; a meeting is what the reply leads to. And ask exactly one thing. Two options stapled together (“worth a call, or shall I send more info?”) give the reader a comfortable way to do neither.

Follow-ups

A follow-up that adds nothing is worse than no follow-up, because it converts a neutral non-response into a mild irritation. “Just bumping this” and “did you see my last email” both belong to that category.

A follow-up worth sending contains something the first email did not: a second signal, a relevant example, a narrower version of the ask. Two follow-ups is a reasonable ceiling for cold outreach, spaced by several days, and every one of them must honour an opt-out immediately and permanently. That is a legal requirement in the UK and EU, not a courtesy. See the PECR guide for what that obliges you to do.

What to do when there is no signal

Some accounts will have nothing current worth opening with. The temptation is to fall back on a generic line so the account does not go to waste. Do not: that is the exact moment a research-based process turns into a template-based one, and the account was never going to reply to the generic version anyway.

Drop the account and revisit it later. A list that shrinks because accounts failed the research test is a list working correctly. See the signals worth waiting for.