Cold email opening lines: how to say why you're reaching out

Search for cold email opening lines and you get lists. Fifty proven openers. Twenty that get replies. Copy, paste, swap the variables.

One of the pages ranking for this today still has an unreplaced placeholder in its own example. Nobody caught it, because nobody using it ever reads the line closely enough to notice. That is the whole problem in one image.

The reason you cannot find a good opening line

An opening line has one job: make it obvious in a sentence why this message came to this person.

That means the line is downstream of a fact. You cannot write it before you have the fact, and once you have the fact the line is nearly automatic. Which is why every list of openers fails. They are the shape of a sentence with the substance removed, and the substance was the part that mattered.

You are not looking for a better line. You are looking for a reason, and the reason has to be found, one person at a time.

The falsifiability test

Before anything else, here is how you know whether what you have is a reason.

Write your opening line. Now take a different person from your list and paste the same line into their message. Does it still make sense?

If it does, it is not a reason. It is a description of a category that both of them happen to belong to.

This test is unforgiving and that is the point. Most personalisation fails it instantly. Mentioning their industry fails. Mentioning their company size fails. Mentioning that they are in a role that typically owns this problem fails. All of those are true of hundreds of people.

What passes: something this specific person said, did, or published, on a date, that you can point at.

The procedure for finding one

Not a list of places to look. An order of operations, because the order is what makes it fast.

1. Start with what they published, most recent first. Posts, comments, talks, podcast appearances, anything where they chose the words. You are looking for a problem stated in their own language, not a topic they discussed.
2. Stop at the first thing that is a problem, not a topic. Someone sharing an article about a subject is a topic. Someone describing something that is not working is a problem. Only the second is a reason.
3. If nothing in the last three months, check what changed around them. New role, new responsibility, a change in what their team owns. This gets you an inferred reason, which is a real Tier 2 reason in the sense described in how to prioritize sales leads.
4. If still nothing, stop. Do not proceed to the part where you invent something. You have learned something true: this person is not currently showing the problem. Put them on the list to re-check.

That last step is the one nobody does, and it is the highest-value step in the procedure.

What to do when there is no reason

Then do not email them.

This sounds like advice that reduces your pipeline. It does the opposite, because of what a generic message actually costs.

You get one first message to a given person. Spending it on something interchangeable does not just fail to work, it converts a future warm contact into someone who has already filed you under noise. When a genuine reason appears three months later, you are no longer a new sender. You are the person who sent the thing they ignored.

Not sending is not a lost opportunity. It is a preserved one.

Four openers that fail the test

Worth being concrete, because these all feel personalised while writing them.

The last one is the interesting failure, because it is the one good salespeople reach for and the one automation produces best.

A company announcement is a fact about a company. Everyone there shares it, so the message survives a swap between any two employees. It becomes a real reason only when you connect it to something specific about this person's remit, and at that point the announcement is the setup, not the reason.

Where to look when they publish nothing

A large share of good buyers publish nothing at all. This is normal and it is not a signal about them.

Three places still work, in descending order of strength.

What they said somewhere that is not a feed. Conference talks, podcast appearances, webinars, industry forums, community threads. People who never post will still speak for forty minutes about their job on someone else's channel.

What their team publishes. A person who is silent often runs a function that is not. Job specifications, internal-facing documentation, and the writing of people who report to them all describe the shape of the work. This is inference, so it belongs in the situational tier and the message should ask rather than assert.

What changed in their remit. A shift in what they own is specific to them, and is usually visible even when the person is not. Again, inference, so ask.

If all three come back empty, you have a genuine answer: this person is not currently showing the problem. That is the point where the correct move is to keep the name and re-check, not to write something generic. How to prioritize sales leads covers where those names sit in the meantime.

Writing the line once you have the reason

The line is close to automatic. Three rules keep it from going wrong.

Lead with their thing, not your reaction to it. The subject of the sentence should be the problem they described, not the fact that you read about it.

Do not announce the research. Saying you noticed something makes the looking into the topic. Engaging with the substance makes the problem the topic, which is where you want the conversation.

Do not quote them back at themselves. Quotation marks around a fragment of someone's own post reads as a mail merge even when it is not, because that is what mail merges do. Refer to what they said in your own words. It shows you understood it rather than captured it.

Then the second sentence connects their thing to the reason you specifically are writing, and the third asks for something small. That is the whole message. It is short because the reason is doing the work.

What good looks like, structurally

Notice there is no room for a paragraph about your company. If the reason is real, the message is three sentences. If you need six, it is usually because the reason is weak and the length is compensating.

How to tell whether it is working

Not by reply rate alone, at least not at first.

The faster signal is the shape of the replies. Specific openers produce specific responses, including specific rejections. Someone telling you why they are not the right person is a sign the message was read and understood. Silence is a sign it was filed.

Track one number over time: the share of your sends where you could write a true, non-transferable reason. That share is the thing you actually control, and everything else follows it. When it goes up and reply rate does not, you have a different problem, and why your outbound stopped working walks through which one.