cyberneticlibrary

Optimize cold outbound to 90/100 quality

outbound-engineskillsetup L32,362
ericosiu/ai-marketing-skills

Causal-lift measurements

cold-email-strategy-33pp vs no-skill baselinewith-skill 57% · baseline 90%

Measured by running the task with and without this artifact, K=5, graded by deterministic checks — no LLM judging.

What it does

Optimize cold outbound campaigns via expert panel scoring to 90/100 quality

Best for

B2B outbound teams running Instantly who want to science their way to higher reply rates instead of guessing, and are willing to iterate email copy through 3-5 scoring rounds until panelists give it 90/100.

Inputs
  • · Instantly API key (optional, for account audit)
  • · ICP definition (titles, industries, company size, revenue floor)
  • · Business context (what you sell, primary offer, real URLs, proof points)
  • · Existing email sequences (if running campaigns already)
  • · Historical outbound data (optional spreadsheet)
Outputs
  • · Infrastructure audit (active campaigns, warmup scores, domain inventory, gaps)
  • · ICP definition document (confirmed targeting parameters)
  • · Expert panel scoring rounds (all 10 panelists, individual scores, aggregate, weaknesses)
  • · Final email copy (all sequences in Instantly-ready format)
  • · Implementation plan (step-by-step setup)
  • · Capacity math (accounts × daily send rate = pipeline projections)
Requires
  • · Instantly.ai (API for campaign audit and implementation)
  • · Expert panel framework (10 defined panelists, scoring criteria per panelist)
Preconditions
  • · ICP must be defined or collected upfront
  • · Business context (what you sell, offer types)
  • · Willingness to iterate copy until 90/100 reached (non-negotiable)
Failure modes
  • · Generic copy never reaches 90/100, loop becomes infinite
  • · Warmup scores <80 or <14 days = emails go to spam
  • · No capacity math means you overpromise pipeline numbers
  • · Expert panel not customized to your vertical → generic feedback
Trust signals
  • · Expert panel framework (10 named panelists, defined scoring lens per panelist)
  • · Non-negotiable 90/100 target with brutally honest feedback
  • · All scoring rounds shown in output (iteration trail is the value)
  • · Capacity math links infrastructure (accounts, warmup, daily limit) to pipeline output
  • · Instantly API integration (audit existing campaigns, not theoretical)