Outbound
Email Is Getting Worse and I Am Part of the Reason
Reply rates are falling, inbox providers are getting stricter, and everyone sending cold email helped cause it.

Tim Burnham
Founder & CEO
August 6, 2026
I send cold email. So I am part of this.
That is worth saying before anything else, because most posts about email getting worse are written by someone selling you a way to send more of it.
Someone put it to me plainly a while back. We are all complicit.
What is actually happening
A campaign we watched went from about a 0.5% reply rate to about 0.2% inside one send cycle. Nothing in the copy changed.
That is not a copywriting problem. That is the channel moving underneath the campaign.
Two things are pushing on it at once.
More email is being sent, because writing a decent looking email now costs almost nothing. Anyone can produce 5,000 of them before lunch.
And inbox providers are responding to that. They do not need to detect that a message was written by AI. They watch what people do with it. Ignored, deleted, marked as spam. Enough of that and your domain quietly stops arriving.
You usually do not get told. Your numbers just get worse and you assume it was the subject line.
The part nobody likes
Every one of us sending volume is training the filter.
The person who replies "remove me" and gets four more emails is not just annoyed at that sender. They are slightly more likely to mark the next cold email as spam, including yours.
The channel is a shared resource and it is being used up.
I do not have a clean answer to that. I still send. I just send less, at people I have an actual reason to contact, and I try not to make the problem worse than I have to.
Turn off open tracking. It adds a tracking pixel that hurts deliverability and measures a number that stopped meaning anything years ago. Watch reply rate instead. Second, stop sending cold email from your main company domain. Use separate sending domains, so a bad month does not take your real email with it.
Why specificity is the thing left
Volume used to be a lever you could pull. Send twice as much, get twice as many replies.
That lever is broken. Sending twice as much now often produces fewer replies, because the extra volume damages the delivery of everything else you send.
So the only input left that still moves the number is how well the message fits the person.
Not personalization tokens. Their first name in the subject line was solved and then ruined ten years ago.
Fit means the message answers why you, why now, and why relevant, using context that only you have. A conversation you had. A thing they published. A change at their company you noticed and can name.
The two best performing outbound workflows I have seen recently both worked for a boring reason. The people receiving them had already interacted with the sender. One had asked to be contacted in the autumn. One had been referred by a colleague.
No clever writing involved. Just context nobody else could have used.
What I would do this quarter
Cut your list. If you cannot say why a person is on it, take them off.
Set a floor. A 0.5% positive reply rate on cold email is a reasonable line. Below it, the problem is almost never the copy. It is the audience or the infrastructure, and you will waste a month rewriting subject lines instead of fixing either.
Check where your email actually lands. Send one to a few real inboxes across different providers and look in the spam folder yourself.
Then send less.
I might be wrong about how long this holds. My honest read is that the channel keeps degrading for people sending volume, and stays fine for people sending 30 emails a week to people they can describe.
Which is fine, if you were already doing the second thing.
Working on something like this? Tell us what's slow and we'll tell you what we'd build.
