I want to confess something that every founder who has done their own outbound will recognize with a small wince. For about two years, my entire follow-up strategy was the phrase 'just circling back.' Email one was always fine — I'd put real thought into it, researched the company, written something personal. Then emails two through five degraded into a sad little loop of 'just circling back on the below,' 'wanted to bump this up your inbox,' and 'following up one last time' before I quietly gave up and told myself they probably weren't a fit anyway. My reply rate across the whole sequence sat around 2%, a number I had decided was simply the natural law of cold email, like gravity or Mondays.
It is not the natural law of cold email. It's the natural law of lazy cold email, and in this story I am unambiguously the lazy one. The thing that finally moved the number — and it moved a lot — was the Cold Outbound Email Sequence — Five Touches That Don't Sound Like a Robot prompt, largely because of one ruthless design choice baked into it: a hard list of banned phrases that includes, by name, the exact verbal crutches I had been leaning on for two years. The first time I ran it, I felt personally attacked, which is usually a sign a tool is working.
The trigger event does the heavy lifting
The prompt flatly refuses to let email one pitch anything before it has earned the open. It demands a real, specific trigger event in the very first line — a funding round, a sudden hiring spree, a product launch, a leadership change, something that proves to the reader you noticed them as a company rather than as a row in a list you bought. I fed it our product one-liner, the customer type, and a trigger: this particular prospect had just posted three open operations roles in a week, which screams 'scaling pain' to anyone paying attention. Grok 4.3 opened email one with a sharp line about the hiring spree and tied it directly to the specific problem our product solves for teams in exactly that moment. It read like I had genuinely studied their company, because the prompt had forced me to actually give it the proof that I had.
That's the subtle thing about this prompt — it can't manufacture relevance, so it makes you supply it. The personalization isn't a trick the AI performs; it's a requirement the AI enforces on you before it'll write a word. That constraint is the entire reason the emails don't sound like a robot wrote them.
Five angles, not five escalating nags
The real upgrade over my old approach was variety with intent. Instead of five emails that all say the same thing with steadily rising desperation, the sequence rotates through five distinct angles: curiosity, a proof point, a genuine question worth answering, a short and specific customer story, and finally a clean breakup. Each email is capped at 90 words and carries exactly one idea, because a busy person scanning on their phone will not read your second idea. Email five — the breakup — assumes the answer is no, leaves the door politely open, and guilt-trips absolutely nobody. That breakup email alone now drives more replies than my old emails two, three, and four combined, because nothing on earth makes a busy person respond faster than the sincere offer to stop emailing them.
- Reply rate climbed from roughly 2% to roughly 6.5% across about 400 sends over two months — more than triple
- The breakup email (#5) became the single highest-reply message in the entire sequence, which still amuses me
- The 'why this works' line the prompt adds under each draft taught me more about outbound psychology than any paid course I'd sat through
- I moved every sequence's send window to mid-morning, Tuesday through Thursday, per the prompt's closing recommendation, and opens ticked up too
The honest caveat about scale
Here is the thing I'd want a friend to tell me before I got excited: this is not a spray-ten-thousand-emails machine, and you will ruin it if you use it like one. The entire reason it works is the personalized trigger event, which means you have to actually research the prospect with your own eyes before you run it. Garbage trigger in, robotic email out — the prompt is only as personal as the input you feed it. So I run it one prospect at a time, exclusively for accounts I genuinely want to land, and for those high-value targets the lift is real, repeatable, and the emails finally stopped making me cringe as I hit send. Trading volume for relevance was the best outbound decision I've made.
Grab the Cold Outbound Email Sequence — Five Touches That Don't Sound Like a Robot prompt on Prompt Dock and run it on one prospect you'd genuinely love to land. Find a real trigger, feed it in honestly, and never type 'just circling back' again as long as you live. When the reply finally lands and you need a sharp one-liner to pitch back fast, I lean on my Value Proposition Rewriter — Kill the Jargon, Keep the Punch prompt to say it plainly.