Cold Email Copy Kills Less Than You Think

Here's What Caught My Eye on LinkedIn This Week

I monitor cold email conversations constantly. Not because I have nothing better to do, but because after sending millions of emails personally and helping over 14,000 entrepreneurs book more than 500,000 sales meetings, I can tell within about ten seconds whether someone posting about cold email has actually done it at scale or is just theorizing.

This week had some genuinely useful stuff mixed in with the usual noise. A few posts made me stop scrolling. A few made me want to reply publicly. Here's the breakdown.

The Infrastructure vs. Copy Debate Is Finally Getting Louder

Multiple posts this week touched on the same core tension. People blame their copy. The real problem is usually upstream.

This is exactly right and I wish more people led with this framing. The order matters more than most people admit. I have watched teams spend weeks perfecting subject lines while their domain reputation was destroyed and 60% of their emails were going to spam before anyone could read the brilliant copy they wrote.

The sequence Jérémy outlines, targeting first, then infrastructure, then messaging, is the same order we teach. When I built my first agency to $600,000 in annual recurring revenue in 60 days using cold email, we were not obsessing over word choice. We were obsessing over who we were contacting and making sure the emails actually arrived. The copy was simple. The targeting and deliverability were locked in.

If you want a complete look at how to set up the technical side, our cold email tech stack guide walks through the infrastructure piece in detail. Tools like Instantly and Smartlead have made a lot of this easier, but you still need to understand what you are setting up and why.

The Offer Problem Nobody Wants to Admit

The diagnosis here is solid. Weak offer, generic targeting, no reason to reply. That list is accurate. I have reviewed hundreds of cold email campaigns where people were convinced the subject line was the problem or the opener was too formal, and when you actually looked at what they were asking for, the offer itself was vague and risk-heavy for the prospect.

"Let's hop on a 30-minute call to explore synergies" is not an offer. That is a sentence designed to make a stranger do work for you with no upside visible to them.

Where I pump the brakes on Conrad's post is the comment section, which is full of people commenting "HORMOZI" to get a GPT. Using a GPT trained on someone else's framework to write your cold email offer is not how you develop the skill. It is a shortcut that will produce average output at scale. The signal-to-noise problem in cold email right now is largely caused by too many people using the same templates and tools. If everyone is running the same Hormozi-style GPT prompt, the output all sounds identical.

Build the muscle yourself first. Then automate.

The Best Cold Email Take This Week Came From Someone Pointing Out Bad Copy

Adam is making a point I have made a hundred times. The research failure here is not subtle. Emailing someone with 150,000 LinkedIn followers to ask if they have "ever considered" podcast appearances is the kind of mistake that happens when you build a list based on job title and company size alone, without actually looking at who you are contacting.

The fake claim is what kills it completely. "Podcast placements add six figures a month" is not a stat any serious buyer believes. The moment you include a claim that the reader knows is false, the entire email is dead. Your credibility does not recover from that within the same message.

The Churchill quote at the bottom is a separate layer of confusion that honestly deserves its own post.

Adam ends with "has anyone tried writing better copy?" which is funny but also honest. The bar really is that low. That is actually an opportunity if you are willing to do the work.

Subject Lines: Someone Finally Shared a Real Example With Real Numbers

This is one of the best subject line posts I have seen in a while because it gives you the actual psychological mechanism, not just a list of "tips."

The open loop principle is what makes this work. When someone sees "Pete, I'd like to buy LeftClick" in their inbox, their brain cannot move on until it resolves the question of who is writing and whether this is real. That unresolved tension drives the click.

The nickname trick is also underused and genuinely effective. Most automated outreach tools default to the formal name from the database. If you know your prospect goes by a shortened version and you use that, you have already created a pattern interrupt before they have read a single word of your email body.

The two-trigger framework Nick lays out, money signal or familiarity signal, is how I think about subject lines too. If your subject line does not trigger one of those two responses, it is neutral at best. Neutral subject lines get ignored. We go deep on this in our subject line breakdown if you want more examples with open rate data behind them.

The 19-Word Email With an 18% Positive Reply Rate

18% positive reply rate on cold email is not normal. For context, most people running cold email at volume are celebrating when they hit 3-5% positive reply rates. 18% means something in this sequence is working unusually well.

What Matt has figured out is the permission ask, which changes the dynamic completely. Instead of delivering a pitch the prospect did not request, he is asking if they want to receive value. That single reframe shifts the prospect from passive recipient to active participant. They said yes. Now they are invested in watching the video you send them.

The specific number, 58 seconds, also does real work. "A quick video" is vague. "58 seconds" tells the prospect exactly what commitment they are making and signals that you have thought about their time. That specificity builds trust before the video even loads.

This is a variation of the PC formula I use. You identify the pain, you make a targeted ask, and you give them one clear next step. The permission layer Matt adds on top of that is what pushes the reply rate into exceptional territory.

Cold Email Math That Most People Skip

This post did not get the engagement it deserved. 23 likes is too low for something this practically useful.

What Sam is doing here is treating cold email like the math problem it actually is. Most people approach cold email like it is magic, something you either have a feel for or you don't. It is not. It is a funnel with inputs and outputs. If you know your book rate, you can reverse-engineer exactly how much infrastructure you need to hit your meeting targets.

The specific gap he identifies is the one I see constantly. Someone sets up 10 inboxes, sends for three weeks, gets two meetings, decides cold email does not work. But they needed 73 inboxes to hit their goal. They had 10. The math was never going to work regardless of how good the copy was.

I have sent well over 60 emails in a single day in early experiments and booked 18 meetings from 60 contacts. But I also know those numbers do not hold at scale without proper infrastructure behind them. The math matters. Build the infrastructure before you judge the channel.

The "Cold Email Is Dead" Crowd Got Called Out Properly

95% inbox rate as the baseline expectation is right. That used to be a stretch goal. Now if you are below that number you are not really running a cold email program, you are running a domain destruction program.

The list Jeana puts together is solid fundamentals. I would add one thing she left off: your list quality is the variable that makes or breaks every element she listed. Perfect copy sent to the wrong ICP still produces zero replies. The research that goes into who you are contacting matters as much as the first line that shows you did research.

Want to see the exact frameworks behind emails that perform at this level? The killer cold email templates we publish walk through real examples with the reasoning behind each structural choice.

A Proof-of-Product Email Worth Breaking Down

This is genuinely smart product marketing and I want to give credit where it is due. The fact that the email landing in primary inbox IS the demo is a product-channel fit that most companies never think to engineer.

There is a broader lesson here that applies beyond inbox warmup tools. Your outreach medium should demonstrate your capability whenever possible. If you are selling video production, send a video. If you are selling copywriting, your email better be the best copy they read that week. If you are selling deliverability, land in primary. The medium carrying the message should prove the message.

Most companies completely miss this. They sell speed and send a slow-loading email. They sell design and send a plain-text pitch that looks like it was written in 2003. Match the medium to the message you are making.

The Post I Disagree With Most This Week

Let me be direct here. "I fed 100M+ cold emails to GPT-5.2" is not a strategy, it is a marketing claim. There is no GPT-5.2 at the time this was written. More importantly, training a model on volume does not mean training it on quality. 100 million cold emails that got no replies is just a very large dataset of what does not work.

The 684 comments on this post are almost entirely people typing "GPT" to get access, which is a lead generation mechanic, not a proof point. The comment volume proves the mechanic works. It does not prove the tool works.

Cold email agencies are not dead. The agencies that were charging for volume and templates without strategy deserved to struggle. The ones building real systems, targeting intelligently, managing deliverability, and writing copy tied to specific pain points for specific ICPs are not going anywhere. A GPT prompt does not replace the judgment those agencies bring.

Be skeptical of any tool that promises to replace the thinking. The thinking is the thing.

The Consistency Point That Everyone Ignores Until Their Domain Gets Burned

This is one of the most underrated deliverability points you will read this week. Email providers are looking at behavioral patterns, not just content. A spike of 1,000 emails on Monday from an inbox that sends nothing the rest of the week reads as spam behavior to the algorithm, even if every single email in that batch was perfectly written and targeted.

Consistency in daily volume is infrastructure discipline. It is not exciting, it does not make a viral LinkedIn post, but it is one of the most important levers you have for protecting long-term deliverability. Ramp slowly, send consistently, and treat your domains like assets, because they are. Once you burn a domain, you do not get the reputation back. You start over.

One More Worth Your Time: Victor Sankin's Remove-the-Name Test

The remove-the-name test is the single most useful self-editing exercise I have seen shared this week. Try it on your own emails right now. Remove your company name. Does the email still make a clear, specific case for why this person should respond? If not, you are writing a brand awareness email disguised as a cold outreach email, and those are two completely different things.

Victor's point about follow-up being part of the design, not an awkward afterthought, is also critical. The best cold email sequences I have built treat follow-ups as a deliberate series of new angles on the same conversation, not repetitions of the same pitch with "just following up" tacked on. Each touch should add something. A new data point, a relevant case study, a different framing of the pain. If your follow-up does not add value, it is just noise.

We have a full breakdown of how to structure follow-up sequences that actually move conversations forward at our cold email follow-up guide.

The Actual Takeaway From This Week

If there is one thread connecting the best posts this week it is this: the lever most people are pulling, copywriting, is the third most important lever, not the first.

The order is targeting, then deliverability infrastructure, then message. Most people skip to the end and then spend weeks optimizing the thing that matters least when the first two are broken.

Get your list tight. Verify your emails before sending using a tool like ScraperCity's email validator so you are not burning your sender reputation on bad addresses. Set up your technical infrastructure properly. Send consistently. Then, and only then, does investing in better copy actually pay off.

The cold email channel is not broken. The approach most people take to it is. Fix the order of operations and the results follow.

If you want the scripts we use at the copy stage once the infrastructure is locked in, the top 5 cold email scripts are a good starting point. Real templates, real numbers behind them, no generic filler.