Cold email follow up: what to say instead of checking in

I have sent "just checking in" more times than I want to count, and it has never once worked. Not because the timing was wrong. Because the sentence has no reason to exist in it, and the person reading it knows that before they finish the clause.

Here is one of mine, three days after a first message that got no reply. The product and the prospect are the same stand-in from before: an AI tool that writes SEO articles, Dana, Head of Content, publishes two posts a week, none of them ranking.

Hi Dana, just wanted to follow up on my previous email. Not sure if you had a chance to see it, but I'd love to hear your thoughts whenever you get a moment. Let me know if you have any questions!

Four sentences, zero information. Read it as Dana, who has already ignored the first one.

Why "just checking in" is the sentence that gives you away

It tells Dana exactly one thing: I sent a message, nothing happened, and I am back with nothing new to say. The words are polite. The function is a nudge, and nudges without content read as pressure, not follow-up.

I wrote that line because the alternative felt worse. Silence looked like giving up, and I did not have anything new to add three days later, so I sent the closest thing to nothing that still counted as an email. That is the actual reason "just checking in" exists as a category of message: it is what you send when you have decided you should follow up but have not decided what to say.

The tell is not the phrase itself. It is what surrounds it. "Not sure if you had a chance to see it" assumes the read receipt of a message she may have opened and closed in two seconds, already having decided. "Let me know if you have any questions" outsources the next move to her, for a pitch she has not agreed is worth questions about. Every clause in that email is there to avoid saying the one thing it is actually about: I want you to reply.

How many follow ups should you actually send

Search this and you get a different number depending on which blog wrote it. Some say two or three. Some say five to seven. A few say follow up until you get a real no, which in practice means indefinitely.

The range is wide because the question is wrong. The number of follow-ups that works is not a fixed count, it is a function of whether each one adds something the last one did not. Three follow-ups that each say something new outperform two that say the same thing twice. A single follow-up that repeats the ask with different punctuation is already one too many.

So the plain answer: three to four touches total, first message included, and every one after the first has to earn its place by adding a fact, a resource, or a narrower ask that the previous message did not have. The moment you are about to send a follow-up and the only thing that changed is the date, stop. That is not persistence, that is the same email twice.

Stopping earlier than that is its own mistake. Dana might be traveling, might have three hundred unread messages, might have meant to reply and lost the thread in a Tuesday. One follow-up that adds nothing is forgivable, expected even. It is the second and third one that need to be different messages, not the same one resent.

What to say instead of checking in

Replace the nudge with one of three things: new information, a narrower ask, or a reason to reply that has a deadline attached. All three give Dana something to react to that was not in the first message.

New information looks like a fact you learned since you sent the original. Not "I noticed your site" as a generic observation, an actual finding: you searched one of her target keywords and she is on page two behind three competitors. That is new, checkable, and specific to her.

A narrower ask replaces "let's hop on a call" with something that takes fifteen seconds to answer. Not "any thoughts?" but "worth a look, yes or no?" A question with two possible answers gets answered. A question with unlimited possible answers gets skipped, because replying to it is its own small project.

A deadline works when it is real. "I'm holding a slot for one more example like this before Friday" only works if you actually mean it, and it only works once. Manufacturing urgency you do not have is the same failure mode as "just checking in," dressed up: words performing pressure instead of creating a reason.

The follow up that got ignored, and the rewrite

Same situation, three days later, no reply to the first message. Here is what I sent the second time, after I stopped writing "just checking in" as my default:

Dana, I looked up "content calendar template" since I sent this. You're on page two, three spots behind [competitor]. I wrote the version that would sit above them and I'll send it if you want it. Yes or no is fine, no explanation needed.

Three sentences. No apology for following up, no assumption about whether she read the first one.

Line one is a fact she can verify in the time it takes to open a new tab. It is not "I did more research," it is the research, stated. That is the difference between telling someone you did something and showing them what it produced.

Line two turns the follow-up into an offer instead of a reminder. I am not asking whether she remembers me, I am telling her what I already made. The first email asked her to imagine the product. This one puts a finished piece of it in front of her.

Line three closes the loop the first draft left open. "Let me know if you have any questions" hands the decision back to her with no shape. "Yes or no is fine" hands it back with exactly two doors, and tells her explicitly that a one-word answer counts as a complete reply. That line alone is why people answer follow-ups: you removed the part where responding felt like it required effort.

When to actually stop following up

Stop after the third silent one, not the first. One unanswered message is nothing, most people's inboxes eat messages that deserved a reply. Two unanswered messages is where a lot of founders quit, and that is usually one message too early, especially if message two added something real.

Three unanswered messages, each one different, each one adding something, is a real answer. Not a no you have to accept gracefully, an absence of yes loud enough to read as one. Sending a fourth at that point does not read as persistent. It reads as not listening, and that is the one thing that can turn silence into an actual no, where before it was just silence.

The exception is a message that gets opened and re-opened without a reply, which most cold email tools will show you. That is not silence, that is interest without an answer yet, and it is worth one more message specifically because you have evidence, not a guess.

The stopping rule changes entirely once she answers. A stated no is not the third silent one, it is a person telling you something the silence never would, and it gets exactly one reply, not three.

Four checks before you send a follow up

Run these before the send button, in this order:

1. Does this message contain something the last one did not?
2. Could Dana answer it in one word?
3. If you removed the words "just," "checking in," or "wanted to follow up," would the message still make sense? If not, it was carrying the whole email.
4. Is this the third one or later? If yes, and nothing changed since the last, do not send it.

The third check catches almost everything. "Just checking in" survives because "just" is doing the work of apologizing for existing in someone's inbox, and once you cut the word you can see there is nothing behind it.

Run all four and the follow-up can still fail for a reason none of them catch: the first email never got opened. That's an earlier problem, with its own fix: how to write a cold email subject line that gets opened.

The habit underneath

The habit is not writing a bad follow-up. It is treating the follow-up as an obligation to perform rather than a second chance to say something worth reading. You already know the first message did not land. The honest move is to ask what you would add if you were pitching Dana for the first time today, with what you know now. Most people skip that question and send the reminder instead, because the reminder is faster to write and does not require you to have learned anything new since Monday. It is faster still if you have a model write it, which is how an empty message ends up sent at scale and why so much outreach reads as generated now.

That gap does not close by memorizing a better subject line. It closes by writing the follow-up, checking whether it actually adds anything, and noticing how often the honest answer is no. Write the draft, then get it read back to you before Dana reads it.

That is what I built Sell a Pen to do: it marks the lines that are carrying nothing, names the habit behind them, and then you write the line again yourself, because the rewrite is the part that sticks. It will not hand you a better sentence, which is the whole point. Start free on a product I made up.