Write cold emails from a researched one-sentence observation, with a reply-style ask and follow-ups
A startup founder shares a five-sentence cold email structure where an AI agent finds one specific observation about each prospect, and reports reply rates rising from about 2% to 5 to 6% on a small SMB test.
Evidence: The author reports this. We have not checked it beyond reading the source.
The business problem
Generic merge-field personalisation gets few replies from small-business prospects, and manual research for each email does not scale.
What was tried
An AI agent searches the web for each company's domain (news, new locations, job postings, review and social updates) and turns the signals into one specific observation. The email has five sentences: the observation, what it implies, what you do, a relevant outcome you produced, and a low-friction ask such as 'worth a quick reply if this is on your radar?' instead of a call booking. Follow-ups go out on day 3 and day 7. Prospect lists come from Apollo and meetings are booked with Calendly.
What was reported (positive)
Over 90 days and 312 emails the author reports a 58% open rate, 5.8% reply rate, 9 meetings and 3 active deals, with about 30% of replies from follow-ups. They say the baseline was about 2% and the formula came from testing roughly 400 emails.
Limitations
The title says 'doubled' but the numbers show roughly two and a half to three times. There is no control group, and the subject lines, follow-ups and call to action changed at the same time as the formula, so the effect of personalisation alone is unknown. The author promotes their own product, does not say whether the tested and reported emails overlap, and does not cover opt-outs, deliverability or email law. The page does not show a year.
What you need
A prospect list source (Apollo is mentioned), a way to gather web signals per company, a sending tool and a booking tool. Costs and skills are not stated.
Sources
- DEV Community (Alex Wu, Anythoughts.ai) ↗ Firsthand write-up, publication date unknown
Source published: unknown. Last reviewed here: October 11, 2026. Spot a mistake? Tell us.
Tools in this workflow
- Apollo: Prospect listsTry Apollo
- Calendly: Meeting bookingTry Calendly
Related workflows
Research sponsors, build a sponsorship CRM and schedule checks with the Manus agent
A content creator used Manus to find companies advertising in AI newsletters, build a web-app CRM for YouTube sponsorships, and schedule recurring lead and trend checks.
Send 500 AI-written cold emails and iterate the prompt with 40-prospect tests, for sponsorship sales
A paid newsletter post says a prompt reverse-engineered from replied-to emails produced 167 replies and 43 meetings from 500 sponsorship cold emails, with each prompt version tested on 40 prospects.
Cut long podcasts into vertical clips with Whisper, Claude scoring, YOLO reframing and n8n
An architecture write-up of an in-house media tool that transcribes podcasts with Whisper, has Claude score the best moments for each brand's audience, reframes video to vertical, and keeps humans in the loop.
Run a scheduled n8n workflow that scrapes leads, checks their GitHub bio and sends GPT-4o-mini cold emails
A builder documents a 21-node n8n workflow that scrapes freelance developer leads, validates emails, looks up a GitHub bio, writes a personalised email with GPT-4o-mini and sends it with pauses between messages.