Hey,
So I’ve been using Akram K’s n8n scraping template, the one that doesn’t use any APIs or AI. It’s been working really well for me. I run search queries like “Hamilton + London + Ontario + Dentist” and scrape a bunch of emails from different neighborhoods. I usually run these in batches of 100 or more.
Now I’m trying to build something more advanced on top of that. What I want is a way to train an AI to understand the way I write, my tone, my structure, the kind of language I use when I’m doing cold outreach, follow-ups, or even just reconnecting with people. Once the AI gets a feel for how I write, I’d like to use it to generate email drafts for the contacts I scrape, not to send automatically, just to have them sitting in my Gmail drafts so I can review and send them manually.
I also want the system to be smart enough to recognize if I’ve already contacted someone before, and in that case, write a follow-up instead of a first-time message. It should also be able to check whether that person replied or not, and if they haven’t replied after something like six days, it should generate a follow-up draft for me.
Ideally, I’d like to run this whole setup manually whenever I’m ready, so every time I execute it, it checks everything: has this person already been emailed, have they replied, is it time for a follow-up, and then it just builds the drafts accordingly.
On top of that, I’d need some sort of memory system built in so it doesn’t keep generating emails for the same people if they’ve already been contacted or if we’re in the middle of a follow-up sequence.
I’ve watched around 20 hours of tutorials on n8n and I’m getting the hang of it, but to be honest I still don’t feel confident that I know the best way to structure something like this. There’s a lot going on here and I want to do it right.
What would you say is the best way to approach building something like this in n8n?
Appreciate your time.