Case study 5

Scoring 700 leads before outreach for $0.09

Romàn from GojiberryAI gave JEV 700 high-intent leads with personalized outreach messages. In 40 seconds it predicted how each message would perform, assigned a confidence score, and detected lead-message mismatches, for $0.09.

Original post "JEV is INSANE. We gave it 700 high-intent leads and personalised outreach messages. In 40 seconds, it predicted how each message would perform, assigned a confidence score and detected lead-message mismatches. All for just $0.09." Read the original post on X

The numbers

metricreported
leads plus messages scored700
time40 seconds
cost$0.09
what came backpredicted performance, confidence score per message, lead-message mismatch flags
reach426,118 impressions, 3,268 likes, 4,020 bookmarks

Why JEV is the right model here

Outreach at volume is a scoring problem. Every lead and every message needs the same judgment applied: does this fit, is this message right for this person, is it worth sending. Doing that with a generative model means paying per message and reading essays before acting.

Scoring first flips the order. You decide what gets sent before anything is sent, so nothing is spent on messages that were never going to land.

How to copy it

  1. Write your fit criteria as a fixed scale, 0 to 100 against your ideal customer.
  2. Score every lead and message before outreach starts.
  3. Send only what clears the bar, and flag mismatches instead of sending them.

Want this set up for your business?

We turn these workflows into working marketing and sales systems.

Book a call