agentleFS
Sign inStart free
prosyo-aiprosyo-ai/gtm-skill/skills/campaign-performance-analyzer/SKILL.md

campaign-performance-analyzer

Skill13 starsChanged 22 days agoMITView on GitHub

Use this skill when you need to understand how outbound campaigns are performing and what to change: \"analyse my campaign results\", \"why is this sequence underperforming\", \"compare my campaigns\", \"weekly outbound report\". Builds the funnel from contacts to meetings for each campaign, step, segment, sender and channel, benchmarks it, finds the stage where performance breaks, separates list, deliverability, copy and follow-up problems, and recommends the next test. Produces a short report a team can act on.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this weekly or at the end of a campaign.

    line 16
  2. Distinctive:

    Read references/outbound-benchmarks.md for typical ranges by channel and what each gap usually means.

    line 30
  3. Distinctive:

    Every conclusion states its sample size.

    line 65
  4. Distinctive:

    MUST report sample sizes and flag anything under 100 contacts as directional.

    line 71
  5. Distinctive:

    MUST check deliverability by sender before judging copy.

    line 72
  6. Distinctive:

    NEVER call a test winner on fewer than 200 contacts per variant.

    line 73

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Use this weekly or at the end of a campaign. It answers three questions: what is working, where does it break, and what do we change next.

Inputs

  • Per campaign: contacts, sent, delivered or bounced, replies, positive replies, meetings booked and held, by step, segment, sender and channel.
  • Opportunities and pipeline created from those meetings.
  • The period.

The play

1. Build the funnel

Contacts → delivered → replied → positive → booked → held → opportunity. Calculate each conversion. Skip open rates as a decision metric; open tracking is unreliable and often turned off for deliverability.

2. Benchmark

Read references/outbound-benchmarks.md for typical ranges by channel and what each gap usually means. Benchmarks are a sanity check, not a target; your own trend matters more.

3. Find the break

The first stage below benchmark is the problem to fix; improving later stages first is wasted effort.

BreakUsually meansLook at
BouncesList qualitySource, verification, catch-alls
Low replies, few negativesNot seen, or not relevantDeliverability by sender; targeting; first line
Replies, few positiveMessage or offer mismatchAngle, persona, ask
Positive, few bookedSlow or weak reply handlingTime to respond, booking flow
Booked, few heldWeak qualification or remindersNo-show rate, lead time
Held, few opportunitiesWrong ICP or poor discoveryFit of meetings, discovery quality

4. Slice it

By step (which email gets the replies), segment (which persona or industry), sender (deliverability or skill), channel. Small slices need samples of 100+ contacts before conclusions.

5. Recommend the next test

One test at a time on the broken stage, with a hypothesis and a success threshold.

Output format

Period: | Campaigns: | Contacts:
Funnel (all): contacts → delivered x% → replied x% → positive x% → booked x% → held x% → opps
Best / worst campaign: <name - why>
The break: <stage - evidence - likely cause>
Slices worth noting: <step, segment, sender findings with sample sizes>
Next test: <hypothesis - change - metric - threshold - sample>
Stop / scale: <campaigns to pause, campaigns to give more volume>

What good looks like

  • Diagnosis follows the funnel order.
  • Every conclusion states its sample size.
  • The report ends in one or two decisions, not twenty observations.
  • The common mistake is optimising subject lines when a third of the emails bounce or land in spam.

Rules

  • MUST report sample sizes and flag anything under 100 contacts as directional.
  • MUST check deliverability by sender before judging copy.
  • NEVER call a test winner on fewer than 200 contacts per variant.

Example prompts

  • "Here's last month's export for 6 campaigns. What's working and what should we change?"
  • "Our reply rate halved this month. Find out why."
  • "Build a weekly outbound report template for our SDR team."

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free