Claude Code · Anthropic
Build a client's cold email campaign from calls and a prospect list
A cold email agency describes how a team member used Claude Code to qualify an event-attendee list for a spice-export client, find verified emails through a provider waterfall, write the sequence and load it into Instantly for review.
What they did
Each client gets a folder with the onboarding form, sales-call transcripts, product details and an ICP matrix, plus market and competitor research Claude Code saves for reuse. For the spice-export campaign it checked each attendee company's website to see what it sells, qualified it against the client's products, and returned separate files for qualified, excluded and needs-review contacts, starting with the first 100. It then ran an email waterfall across several enrichment providers with validation, only for the top tiers, and wrote copy using a skill built on the agency's past campaigns and replies.
How it went
The poster says the whole campaign can be built without their usual list-building tool, and that the copy skill suggested an offer drawn directly from what the client's product does, finding buyers for the prospect's commodity in a market they want to enter.
Worth knowing
The post walks through one demo session; it reports no reply or meeting numbers, and the presenter noted they should have deduplicated before enrichment. The agency sells outbound services, so the post is partly promotional.
Try it with Claude Code
In <client folder>, read the call transcripts, product sheet and <ICP document>. Take the first 100 rows of <prospect CSV>, remove duplicates, visit each company's website and decide whether it fits, writing qualified, excluded and needs-review files with a reason for each. For qualified contacts only, find a work email using <your enrichment tools> and record which source and check it passed. Then draft a three-step email sequence for my review; do not upload or send anything.
We wrote this prompt from the case. Replace the parts in angle brackets with your own details.
What this label means: User reports
Someone using it at work posted about it. Any numbers are theirs, so we write "the poster says".
Real cases of AI agents at work, from every major agent, each linked to where it was first reported. Rewritten in our own words, most with a prompt you can try. These are not Wize customers.