── SKILL
sequence-performance-analyst✓ APPROVED

Find where your sequences leak, stage by stage.

Breaks each cadence down, finds the drop-off, benchmarks it, and hands back specific fixes.

Inside: The stage-by-stage breakdown, the benchmarks, and the fixes.

01 / HOW TO USE

Install it in one line, or paste it in.

1
In Claude Code (one command)
Copy the install line, paste it into your terminal, and restart Claude Code. The skill installs itself to ~/.claude/skills/ and runs automatically when it is relevant.
2
In Claude, ChatGPT, or a Project (no terminal)
Open the file, then upload it to your chat or paste its contents in. A skill is just a markdown file of instructions, so any capable AI can follow it.
New to skills? A skill is a plain-text file that teaches your AI a workflow. Point any capable assistant at it and it follows the steps, on your data.
02 / MAKE IT YOURS

Connect your context. Set it to your motion.

CONNECT YOUR CONTEXT · AND WHY IT HELPS
a sales-engagement or sequencing tool

all of the above pulled live and kept current, the authoritative source for outreach analytics.

One rule either way

outreach metrics come only from the sequencer. Do not pull them from email or a CRM, connected or pasted.

SET IT TO YOUR MOTION

Point the sequencing-tool binding at whatever cadence platform you run. Every benchmark above is a default from outreach research. Replace any of them with your own measured rates once you have volume.

Set thisWhat it isDefault / Example
Sequencing toolThe connector holding your sequences and per-stage metricsAny sales-engagement platform
Reply-rate floorMinimum acceptable reply rate for cold prospecting12%
Reply-rate crisis lineBelow this signals fundamental targeting or messaging problems5%
Consecutive-email flagBack-to-back emails that trigger a channel-change warning3 in a row
Cadence templateProven day-spacing between stagesDay 1371421
Inbound cadenceFaster spacing for inbound leads9 touches over 10 days
Outbound cadenceSlower spacing for outbound leads9 touches over 15 to 27 days
Touch-count targetTouchpoints typically needed to book a meeting8 to 12
Give-up flagStage count below which a sequence quits too early4 to 5 stages
Your channelsThe non-email touches you alternate inLinkedInphoneyour channels
03 / THE FULL SKILL

Everything the skill does, in full.

You are a sequence optimization analyst. Pull sequence data from the connected sales-engagement tool, analyze performance at the stage level, benchmark against proven best practices, and give the rep or manager specific, actionable recommendations. Not just numbers.

── WHAT THIS DOES

This skill reads your email sequences or cadences, breaks each one down stage by stage, finds where prospects drop off, and compares the numbers against known outreach benchmarks. It returns plain-language insights and specific fixes: which subject line to change, where to add a non-email touch, how many stages to add, and which sequences to retire. It works on a single sequence or across all of your active sequences.

── THE METHOD
  1. 1
    1. Scope the analysis. One sequence, all sequences, a time period, or a specific problem.
    1. Pull sequence data. Stages, subject lines, timing, recipient counts, and per-stage metrics. Check daily send volume against limits.
    2. Stage-level analysis. For each stage report open, click, reply, bounce, and the drop-off from the previous stage. The most valuable insight is usually where the drop-off happens.
    3. Benchmark. Reply rate against the 12% floor. Flag 3+ emails in a row without a channel change. Compare timing against the cadence template. Flag sequences that stop at 4 to 5 stages.
    4. Generate specific recommendations. Each one names the exact change. Not "improve your subject lines," but the new subject line to test.
    5. Comparative analysis across sequences. Rank by reply rate, identify what the best one does differently, flag long-running poor performers doing active harm.
── QUALITY GATES
  • Pull sequence metrics only from the connected sequencing tool. Never source outreach numbers from email or a CRM.
  • Lead with the insight, not the number. "You're losing half your prospects at Stage 3" beats "Stage 3 open rate 18%."
  • Every recommendation names a specific change. When suggesting an A/B test, propose the two specific variants.
  • If a sequence performs well, say so and explain what is working so it can be replicated.
── OUTPUT (EXAMPLE)
EXAMPLE OUTPUT
Sequence: Cold Outbound - Operations Leaders
Recipients: 420   Stages: 4   Span: 10 days
Overall reply rate: 6%  (benchmark 12%)   VERDICT: Underperforming

Stage-by-stage
  Stage 1  Day 1    Open 38%   Reply 4%
  Stage 2  Day 3    Open 31%   Reply 3%   (drop-off -18% open)
  Stage 3  Day 6    Open 14%   Reply 2%   (drop-off -55% open)
  Stage 4  Day 10   Open 12%   Reply 1%

Do this
  1. Rewrite the Stage 3 subject line. A/B test "Quick thought on [their
     challenge]" vs "[their industry] teams are seeing this."
  2. Convert Stage 2 from a meeting ask to a value-add.
  3. Insert a non-email touch (LinkedIn, phone) between Stages 2 and 3.
  4. Extend to 7 stages with increasing time gaps.

Sending volume: 145/150 per day. No headroom. Spread sends or prioritize.
── WHERE THE NUMBERS COME FROM

Every threshold is a default from outreach research, not a law. The 12% floor and 5% crisis line are cold-prospecting benchmarks. The Day 1/3/7/14/21 cadence and the inbound vs outbound spacing are proven defaults. The 8 to 12 touches and "persistence lifts conversion" figures are industry data. Re-tune all of them against your own won-deal touch counts when you can.

── EXAMPLE CHAT
EXAMPLE CHAT
You
How's my Cold Outbound sequence doing? Here's the export, 420 recipients, 4 stages.
CLAUDE
Pulled it apart stage by stage. You're losing more than half your prospects at Stage 3. Sequence: Cold Outbound - Operations Leaders Recipients: 420 Stages: 4 Span: 10 days Overall reply rate: 6% (benchmark 12%) VERDICT: Underperforming Stage-by-stage Stage 2 Day 3 Open 31% Reply 3% (drop-off -18% open) Stage 3 Day 6 Open 14% Reply 2% (drop-off -55% open) Do this 1. Rewrite the Stage 3 subject line. A/B test "Quick thought on [their challenge]" vs "[their industry] teams are seeing this." 3. Insert a non-email touch (LinkedIn, phone) between Stages 2 and 3. The lead insight isn't the 6% overall reply rate, it's that Stage 3 is where you lose them. That's a specific, fixable spot, not a vague "your sequence isn't working."
You
Is 6% actually bad, or is that normal for cold outbound?
CLAUDE
It's below the 12% floor for cold prospecting and getting close to the 5% crisis line, so it's not just soft, it's worth fixing now before it gets worse. Sharpener: connect your sequencer directly and I'll pull this fresh every week instead of working off a one-time export, so you catch the Stage 3 drop-off before it costs a whole quarter of replies.
04 / GO FURTHER

Where an operator takes this next.

The one-time pull finds the leak. Here is the version that watches for the next one.

The fix gets found and tested without waiting for the next manual pull.

01
Analyze every active sequence weekly

Schedule a Claude task against your sequencer so the comparative ranking runs itself and flags the worst performer before it burns more list.

02
Catch send-volume risk before it caps you

Connect your sequencing tool so the daily-volume check runs live, not from a stale export.

03
Test the subject-line fix automatically

Push the two A/B variants into your sequencer so the recommended test launches instead of sitting in a doc.

05 / PART OF A BIGGER PLAY

One skill is the on-ramp.

── GET NEW SKILLS IN YOUR INBOX

One operator-built skill a week. Free.