I’ve been working on a project after running into the same problem repeatedly while doing B2B outreach.
Most lead generation tools either dump massive unqualified lists or require hours of manual filtering and personalization. Scraping is easy, but deciding who is actually worth contacting and what to say is where things break down.
The approach I’m experimenting with is:
* Pulling real businesses from public sources (e.g., Google Maps)
* Scoring leads based on signals like reputation, activity, and fit
* Generating outreach that’s tied to those signals instead of generic templates
The goal isn’t more emails sent, but fewer emails sent to the wrong people.
I’m curious if others here have tackled the same issue:
* How do you currently qualify leads at scale?
* What signals have actually correlated with replies or conversions?
* Have you seen ML-based scoring work in practice, or is rule-based still more reliable?
Would appreciate any lessons learned or failure stories.
Schmiedey•1mo ago
Most lead generation tools either dump massive unqualified lists or require hours of manual filtering and personalization. Scraping is easy, but deciding who is actually worth contacting and what to say is where things break down.
The approach I’m experimenting with is:
* Pulling real businesses from public sources (e.g., Google Maps) * Scoring leads based on signals like reputation, activity, and fit * Generating outreach that’s tied to those signals instead of generic templates
The goal isn’t more emails sent, but fewer emails sent to the wrong people.
I’m curious if others here have tackled the same issue:
* How do you currently qualify leads at scale? * What signals have actually correlated with replies or conversions? * Have you seen ML-based scoring work in practice, or is rule-based still more reliable?
Would appreciate any lessons learned or failure stories.