On October 26 at 18:00 I sent 17 cold emails in 61 seconds. Fifteen follow-ups, one breakup, one first touch. Zero replies came back.
That was not my first batch. Since October 10 I had sent 102 cold emails to real businesses. Replies from clients: zero. Three bounced.
I could call that "cold email doesn't work" and move on. But the sends went out, the tracking showed opens were rare, and the silence had a shape to it. So instead of sending more, I spent a week figuring out where exactly each email died.
It turns out there are three places. Every dead email I traced died in one of them, long before the pitch mattered.
1. The address was wrong
This is the quietest killer because nothing tells you it happened. No bounce always arrives. Some dead addresses just swallow the email. Your sequence fires, your dashboard says "sent," and nobody was ever on the other end.
I learned this the slow way. In my first 87 sends, three bounced outright. That was the visible part. The invisible part was the addresses that looked fine — info@, contact@, hello@ — and went to inboxes nobody reads, or to people who left the company two years ago.
The test takes two minutes. Before you send anything, check the address the way a spam filter would: does the domain accept mail at all, does this exact mailbox exist, and is the person behind it still at that company? If you cannot confirm all three, do not send. A send to a dead address is not a neutral event. It is a small ding on your domain's reputation, and those dings add up until real inboxes start filtering you too.
I now verify every address before the first send. Bounces dropped to near zero. That did not get me replies by itself. It stopped me from burning the domain while I fixed the other two problems.
2. The subject earned nothing
A business owner gets dozens of cold emails a day. Most subjects are about the sender: "Quick question for you." "Partnership opportunity." "Following up." The prospect deletes those on reflex, in under a second, without ever seeing the body.
A subject does not need to be clever. It needs to survive the delete reflex. The only subjects that survive are the ones that are clearly about the prospect: their business, their street, their problem, something that could not have been sent to a hundred other companies.
My test: cover the prospect's name and company in your subject line. If the subject still makes sense, it could have been sent to anyone, and it will be treated that way. Rewrite it until it only fits them.
Subjects that fail the test are easy to spot. "Quick question for you." "Partnership opportunity." "Following up." They say nothing about the reader and everything about the sender's hope. A subject that passes sounds almost boring by comparison: it names the business and the one thing you actually saw — their street, their service list, the gap on their site. Boring is fine. Boring and specific beats clever and generic, because specific survives the delete reflex and clever does not.
I used to write subjects about my offer. Now I write subjects about their shop, their town, the thing I actually saw on their site. Open rates are still nothing to brag about. But the emails that do get opened get opened by people who already know the email is for them.
3. The opener was generic
"I came across your company and was impressed by your work." The prospect has read that sentence a hundred times this week. The first line decides whether the next four seconds happen at all. An opener that could be sent to anyone convinces no one.
Personalization is not decoration. It is the proof the email is for them. One specific detail from their site — a service they actually list, a town they serve, a photo on their homepage — does more work than three paragraphs of pitch. If the first line does not contain something you could only know about this business, the reader assumes a mass blast and stops reading. They are usually right.
My test: read your first line out loud and ask whether it would still be true if you sent it to the business next door. If yes, delete it and start with the detail instead.
Run the test on your last 10 sends
You do not need a new tool for this. Take the last ten cold emails you sent and sort each one into the three places. The sort takes ten minutes and it tells you more than any open-rate dashboard.
For each email, ask three questions in order:
- Was the address confirmed — mailbox exists, person still there? If you cannot answer yes, the email died in place one. Count it.
- Would the subject survive if the prospect's name were removed? If it still makes sense for any business, it died in place two.
- Would the first line still be true if sent to the business next door? If yes, it died in place three.
My own last batch sorted like this: most of the early sends died in place one and three at the same time. Generic addresses, generic openers. The subject lines were the least broken part, which surprised me — I had spent the most time on the pitch and almost no time on the address.
One run shows the pattern clearly. I worked through 34 dental clinics once, sent 9 emails, and graded exactly one clinic A-tier — meaning the address was confirmed, the owner was identifiable, and the opener had a detail only that clinic could match. The other eight were B-tier or worse. That is the real ratio when you grade honestly: about one in nine sends is actually built to land. The rest are lottery tickets with your domain's reputation as the stake.
What changed
I stopped sending 200-email batches. Five emails a day, each one verified, each one with a subject and opener that only fits that business. I build a small demo of what I am offering before I write the email, so the prospect sees something real instead of imagining it.
The cadence changed too. The 17-emails-in-61-seconds night taught me that speed is not the bottleneck — the bottleneck is how many of those 17 were built to land. Sending faster just burns the list faster. Five careful sends a day beats seventeen fast ones, because the five do not bounce, do not read as mass mail, and do not cost the domain anything.
I also stopped treating the follow-up as a copy of the first email. A follow-up that adds nothing new — "just bumping this up" — dies in the same place the first one did, only quieter. If the first email died because the opener was generic, bumping it does not fix the opener. Each touch has to carry something the previous one did not: a detail, a demo, a different angle on the same problem.
Has it fixed everything? No. The reply count is still the number I started with, and I am not going to dress that up. What changed is the shape of the silence: no more bounces, no more dead addresses, opens from the right people. The emails now die later, closer to the pitch — which means the next thing to fix is the pitch itself, not the plumbing.
That is the honest math of a cold outreach system built by hand: you fix one place where emails die, then you find the next one.
What is inside the kit
The system behind all of this is five scripts, four SOPs, and twenty prompts — the pieces I actually use, packaged as the LeadGen Agent Kit. No lead subscriptions. No tool stack eating the margin. Just a pipeline: verify the address, earn the open, start with a line that cannot be deleted.
The free starter kit is the first two pieces of that pipeline: the email-gate SOP that verifies any address before you send, and three opener prompts that write first lines a prospect cannot mistake for a mass blast. That is the part that fixes place one and place three. Place two — the subject — is a habit you build by running the test above on every send until the generic subjects embarrass you before you hit send.
Why I publish the misses
This is my 30th article on Dev.to. The 26 I had published by October 31 brought 119 views in total. The best one — the 34-clinics piece — has 32 views. Nobody is going to call that a traffic machine.
I publish the misses anyway, for two reasons. First, the articles are the record. When a send bounces or a subject dies, the note goes in the article the same week, while the numbers are still warm. A system you cannot look back at is a system you cannot fix. Second, the people who do read them are the people the funnel is for: solo founders and freelancers sending outreach by hand and hearing nothing back. They do not need another guru thread about ten-x-ing replies. They need someone to say the reply count is zero, here is where the emails died, and here is the check that caught it.
That is also why the funnel page says the same three things this article says. The article is the diagnosis. The free kit is the first fix. The paid kit is the whole pipeline for the people who want it. No step in that chain promises a reply count I do not have.
If you run the 10-send test tonight, write down the sort. How many died in place one, how many in place two, how many in place three? My guess, from my own batches: place one and three will take most of them. Fix the address first. It is the only one of the three that costs you reputation even when nobody reads a word.
Where to start tonight
If your emails are dying too, start with the first place — the address. I packaged the check I run now as a free starter kit: the email-gate SOP that verifies any address before you send, plus three opener prompts that write first lines a prospect can't mistake for a mass blast.
You can get it here: Get the free starter kit — email-gate SOP + 3 opener prompts
Leave your name and email, and run one of your real campaigns through it tonight. Compare what happens with the sends you were about to fire blind. Then tell me where your emails are dying — bounces, spam, or silence. Your answer shapes what I write next.
Top comments (1)
Point 1 matches what I found from the other side. I scraped 63 small-business websites to see where their real addresses live, and a lot of "wrong address" cases come from the list, not the person: the footer email belongs to the web agency that built the site, or a template leftover like info@mysite.com was never replaced. Another common one: the only working email is on the privacy or complaints page, while the contact page just has a form.
A cheap check before sending: does the address sit on the company's own domain (or is it a Gmail/Outlook address the business clearly uses), and does that domain have mail servers (an MX lookup)? It won't catch a mailbox that silently swallows mail, but it removes the obvious dead ones for free.