mcpField
C
campaign-tiering

campaign-tiering

github.com/victor-shulgaMarketing & growth146 installsUpdated 4d ago
Repository

About campaign-tiering

campaign-tiering is a skill for teams reviewing active outreach campaigns against performance data. It classifies campaigns as scale, optimize, or kill based on interest and reply rates, while excluding samples with too few sends.

Key features

  1. 01Tier campaigns by interest and reply rates
  2. 02Exclude campaigns with fewer than 50 sends
  3. 03Suggest growth actions for scale-tier campaigns
  4. 04Recommend a test for optimize-tier campaigns
  5. 05Identify likely causes for kill-tier campaigns

Use cases

  1. 01Review active outreach campaigns weekly
  2. 02Compare campaign performance across hypotheses
  3. 03Decide which campaigns to pause or expand

MCP support

The workflow pulls campaign metrics from a sending platform via MCP, then calculates interest rates and assigns tiers.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/victor-shulga/outbound-engine-skills --skill campaign-tiering

After installing, type/campaign-tieringin Claude Code to confirm it loaded.

Details

From the project README

Campaign Tiering

Review all running campaigns and sort them into three tiers — scale, optimize, or kill — based on performance data. Do not keep running what isn't working.

What you need

  • Campaign performance data (reply rate, interest rate, open rate)
  • List of active campaigns with hypothesis names
  • Number of contacts sent per campaign (need minimum 50 sends to tier)

Performance thresholds

Tier 1 — Scale: Interest rate above 30% → Add LinkedIn as a second channel. Increase send volume. Expand the contact list.

Tier 2 — Optimize: Interest rate 17–30% → Tweak copywriting. Adjust CTA. Try a new subject line variant. Don't kill yet.

Kill — Under 17% interest rate (or under 1% reply rate) → Pause immediately. Do not add more contacts. Document what failed and why.

Interest rate definition: % of replies that expressed genuine interest in continuing the conversation (not out of office, not unsubscribe, not "not interested").

Process

  1. Pull campaign metrics via MCP from the sending platform
  2. Filter out campaigns with under 50 sends — not enough data to tier
  3. Calculate interest rate for each campaign: (interested replies / total replies) × 100
  4. Assign tier per thresholds above
  5. For Tier 1: suggest specific scale actions
  6. For Tier 2: suggest one specific optimization to test
  7. For Kill: identify the most likely root cause (targeting, copy, angle, or signal)

Output format

Campaign tiering review: [date]
Campaigns evaluated: [n] | Excluded (under 50 sends): [n]

TIER 1 — SCALE ([n] campaigns)
[Campaign name] | Sends: [n] | Reply rate: [n%] | Interest rate: [n%]
→ Scale action: [add LinkedIn / expand list / increase volume]

TIER 2 — OPTIMIZE ([n] campaigns)
[Campaign name] | Sends: [n] | Reply rate: [n%] | Interest rate: [n%]
→ Optimize: [specific suggestion]

KILL ([n] campaigns)
[Campaign name] | Sends: [n] | Reply rate: [n%] | Interest rate: [n%]
→ Root cause: [targeting / copy / angle / signal quality]

SUMMARY
Campaigns to scale: [n] | To optimize: [n] | To kill: [n]
Recommended next step: [one action]

Notes

  • Run tiering weekly — campaigns that looked promising at 20 sends may fail at 100
  • Killing campaigns is good news: you freed budget and attention for what's working
  • If everything is in Tier 2 or Kill, the problem is usually the signal strategy, not the copy
  • Never move a campaign from Kill back to active without changing the hypothesis — same campaign, same result
  • Campaigns with high open rate but low reply rate = copy problem. Low open rate = subject line or deliverability problem.

Comments

Markdown supported