How to structure a cold message that gets a reply

Four parts, in order, under a hundred words. Most cold messages fail on structure rather than wording.

By The Coaching World

Published October 26, 2026 · 6 min read

How to structure a cold message that gets a reply

Four parts, in order

A cold message that works has four components and almost no other content. A reason you are writing to them specifically, a problem stated in their terms, a brief reason to believe you, and a small ask.

Most failing messages contain all four in the wrong order, buried in a fifth element that should not be there.

This article is for anyone rewriting outreach for the third time. It covers what each part does, how long the whole thing should be, the things to remove, and what to change when it does not work.

The opening line

Its only job is to establish that this was written to them rather than to a list.

That does not require elaborate personalisation. A reference to something specific about their situation is enough, and it can come from a category rather than an individual. Being told the message is for people in a particular role facing a particular change is specific enough to pass the test.

What fails is flattery and the appearance of research. Praising a recent post reads as a template with a slot in it, because it is. Prospects have seen hundreds.

Also failing: starting with yourself. Any sentence beginning with your name or your company has spent the only line you had.

The problem statement

The sentence that does the work, and the one most worth time.

It should describe a situation they would recognise, in words they would use. Not a benefit, not a solution, the problem as experienced.

The test: could the recipient have written this sentence about their own week. If it sounds like a description of what you sell, it came from the wrong side of the table.

This is where buyer conversations earn their value. A phrase lifted from someone describing their own situation outperforms anything composed internally, because it contains the detail that invention smooths away.

Keep it to one sentence. Two is already explaining.

The reason to believe

Short, specific, and immediately after the problem.

One line. Something you have done that makes the problem familiar. Not credentials, not a list of clients, not years of experience. Evidence that you have been close to this particular situation.

The common mistake is proportion. A paragraph of credibility in a message this short reads as a pitch, and it answers a question the reader has not asked yet.

If you have nothing yet, say less rather than inventing. An honest message with no proof performs better than one with manufactured authority, which experienced buyers notice.

The ask

Smaller than a meeting, almost always.

A question they can answer in a sentence converts far better than a request for thirty minutes, because it costs nothing and does not commit them to being sold to.

Make it genuinely answerable. Asking how they currently handle something is answerable. Asking whether they would be interested in learning more is not a question, it is a request wearing one.

If you do ask for a call, say what it will cover and how long. Unspecified time is the thing people decline.

Length and what to remove

Under a hundred words. Under eighty is better.

Cut anything that explains your methodology, anything that lists more than one problem, anything that hedges, and any sentence that exists to set up the next one.

Remove the second paragraph about yourself. There should only be one line about you and it belongs where it is.

Remove links in a first message. They look like marketing and some platforms treat them unfavourably.

And remove any sentence you would not say out loud to someone in a corridor. Written outreach drifts into a register nobody uses in speech, and that register is what makes messages feel automated.

What to change when it fails

Change one thing and wait for enough volume to read the result.

If there are no replies at all, the problem statement or the targeting is wrong. Change the problem sentence first, because it is cheaper than rebuilding a list.

If replies arrive but say wrong person, the list is wrong and no wording fixes it.

If replies arrive and stall, the ask is wrong. Make it smaller.

Changing the opening line is where most people start and it is rarely the problem.

Why short wins

The instinct is that a short message cannot convey enough, and the opposite is true for a first contact.

The recipient is deciding in about four seconds whether to keep reading. A long message loses that decision before any of its content is reached, so the extra detail never gets delivered.

Length also signals effort in the wrong direction. A substantial message from a stranger implies a substantial expectation attached, and people decline expectations before they evaluate them.

There is a practical reason too. On most platforms and in most inboxes, a long message is truncated in preview. The part that persuades is frequently the part behind the fold.

The job of a first message is to earn a reply, not to make a case. The case comes later, in a conversation, where you can respond to what they actually say.

Personalisation, honestly

There is a long-running argument about whether to personalise each message, and the answer depends on numbers.

Deep personalisation works and does not scale. Researching each prospect and writing individually produces strong response rates and consumes a day for twenty messages. For a small market or high value clients, that trade is correct.

Segment personalisation scales and works almost as well. One message written precisely for a narrow group reads as specific because the situation it describes genuinely is theirs. This is where most businesses should be.

Token personalisation, where a name and a company are dropped into a template, performs worse than no personalisation at all, because it signals a system while claiming attention.

If you cannot write something true of a whole segment, the segment is too broad. That is a targeting conclusion, not a writing one.

A note on follow-ups

The structure above applies to the first message only.

A second message should do a different job, not repeat the first more politely. Offer something, ask something else, or say something useful with no request attached.

Two or three touches is reasonable. Beyond that, the pattern of silence is itself the answer, and continuing damages the impression you spent the first message making.

One last check before sending anything at scale: read the message aloud. Anything you stumble over, or would never say to a person standing in front of you, comes out. That single pass removes most of what makes outreach sound automated.

Our articles are researched and reviewed by our editors. Read our editorial standards.