── SKILL
strike-zone-analyst✓ APPROVED

Find where the funnel leaks and what to fix.

Finds the leak, the accounts to work next, and where the scoring model is quietly letting you down.

Inside: The funnel-leak diagnosis, the account ranking, and the scoring audit.

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 CRM

cohort dates, stages, scores, contacts, and owners, live.

a product-analytics tool

real product engagement for the Product channel.

SET IT TO YOUR MOTION

Swap in your own scoring model, your own selling methodology, and your own channel and stage names. The method below is written against placeholders so it does not assume any one vendor or taxonomy.

Set thisWhat it isDefault / Example
CRM connectorWhere cohort dates, stages, scores, contacts liveYour CRM
Product-analytics connectorWhere product-usage events liveYour product-analytics tool
ICP-qualification connectorIndependent fit qualifier (optional)Your ICP tool
Community/intent connectorCommunity + buying-committee signals (optional)Your community tool
Enrichment connectorVerified contact data on top-tier accounts (optional)Your enrichment tool
ChannelsYour funnel entry channelsInbound / Outbound / Product
Cohort anchor field per channelThe date field that marks funnel entryA "qualified date" field per channel
Qualification stageThe CRM stage that equals SQLYour "Qualification" stage
Discovery stageThe CRM stage that equals SQOYour "Discovery" stage
New-business filterThe opportunity-type filter that excludes renewalsOpportunity type = New Business
Fit-score fieldThe model score you are auditingYour scoring model's fit score
Score floor valueThe value your engagement score collapses to when usage is sparseYour model's floor score
Conversion benchmarksPer-gate target conversion ratesYour trailing baseline
Average deal sizeUsed to dollarize leverage pointsYour current ASP
03 / THE FULL SKILL

Everything the skill does, in full.

── WHAT THIS DOES

This is your diagnostic partner for understanding a sales funnel: why it is converting or not, which accounts to work next, and where the scoring model is letting you down. A scoring dashboard tells you what is happening. This skill helps you understand why, what to do about it, and which accounts deserve the next sprint. It works across three modes that share one six-gate funnel framework, one multi-source signal stack, and the same cohort-anchored math.

── THE METHOD
  1. 1
    The six funnel gates

    Strike zone is the sequence of conversion gates from a new lead to closed revenue. For each cohort and channel, six gates matter:

    1. Cohort created (a qualified-account date fires) to Meeting booked.
    2. Meeting booked to Meeting completed (show rate).
    3. Meeting completed to SQL (your Qualification stage).
    4. SQL to SQO (your Discovery stage).
    5. SQO to Closed Won.
    6. Closed Won to average deal size and cycle time.

    Each gate has a different failure mode and a different intervention. Misdiagnosing the gate wastes weeks fixing the wrong thing.

  2. 2
    Mode 1: Funnel Diagnosis

    Default mode is conversational. Work interactively. Do not dump a full report unprompted.

    1. Anchor on a channel and a cohort window. Ask what they want to diagnose.
    2. Pull live data from the CRM. Always run queries. Do not analyze from memory.
    3. Find the leaky gate. Compare each gate to the channel's trailing baseline and to the other channels. The leaky gate has the largest negative gap.
    4. Diagnose with the pattern framework. Each leaky gate has two to four known failure modes. Identify which fits.
    5. Recommend a specific action. Named: a person, a list of accounts, a process change, a target.
    6. Offer the written write-up only if asked.
  3. 3
    Mode 2: Sprint Planning

    Converts the funnel from a diagnostic into an execution backlog.

    1. Define the cohort. Default: all accounts whose qualified-account date falls in the target window.
    2. Multi-source enrich every account in parallel: CRM firmographics and stage, ICP-tool fit, product-analytics engagement, community signals, and enrichment contacts on the top tier only (the paid step, check credits first).
    3. Compute composite scores. Tier into Hot, Warm, Watch, Disqualify.
    4. Build the spreadsheet: Sprint Ranking, Top Buying Committee, Scoring Rubric, Cohort Summary, Scoring Gaps.
    5. Hand off. Reps work the buying-committee tab top-down.

    Cost guardrails. The enrichment tool is the only paid source. Cap spend per run unless approved.

  4. 4
    Mode 3: Scoring Audit

    When the independent qualifier, product analytics, community signals, and CRM disagree with your scoring model, the disagreement is the finding.

    • Model False-Negatives: the engagement score sits at its floor value, but product analytics show real sustained team adoption.
    • Model-Missed ICPs: an independent qualifier scores a strong fit, but the model's account-fit score is low or null. A firmographic miss, not an engagement miss.

    Common scoring failure modes: score-floor collapse, "no score" read as low fit, no buying-title weighting, no trend velocity, no ICP penetration ratio.

  5. 5
    Action format (all modes)

    Every recommendation needs three parts: Who (named person or role), What (a specific change), By when (a date or sprint). If you cannot fill all three, diagnose deeper.

── QUALITY GATES
  • Use your own selling methodology consistently. Do not introduce a framework the team does not run.
  • Lead with conversion, not volume.
  • Cohort math, not snapshot math. Anchor the cohort by entry date and trace it forward.
  • Honest caveats. If a cohort is too young or a field is not groupable, say so.
  • Use the audience's own taxonomy. Keep expansion as its own motion, not a funnel channel.
  • If a channel-source field is a formula field, filter on it but do not group by it.
  • Filter renewals out of new-business math.
  • If a query needs a field you cannot confirm exists, fail loud and ask. Never invent field names.
── OUTPUT (EXAMPLE)
EXAMPLE OUTPUT
STRIKE-ZONE FUNNEL DIAGNOSIS  ·  Trailing 90 days

OUTBOUND COHORT
  Meeting Set:  47
  to SQL:       16  (34%, BELOW 55% target)
  to SQO:        9  (56%, strong)
  to Closed Won: 4  (44%, strong)
  Velocity: avg 82 days meeting to close
  Diagnosis: Leakage at MEETING to SQL. Getting meetings, not qualifying them in.

LEVERAGE POINT (dollarized):
  Outbound meeting-to-SQL +10% = 5 more SQLs per quarter
  At ASP $22K that is +$110K pipeline + 2 more Closed Won
  = +$44K ARR per quarter at current win rate

WHAT MOVES THE NUMBER:
  1. Listen to 5 outbound meeting recordings, find the qualification gap
  2. Test a qualification-first meeting structure on the next 10 outbound meetings
  3. Compare Inbound vs Outbound conversion at the SQL stage to find the delta
── WHERE THE NUMBERS COME FROM

The benchmarks, target rates, the ASP, and the floor score in the example are defaults. Re-tune them to your own data. Set each gate's target from your channel's trailing baseline. Replace the ASP with your current average deal size before dollarizing. Pick a cohort window old enough that the cohort has had time to convert, and report sample size when it is young.

── EXAMPLE CHAT
EXAMPLE CHAT
You
Outbound felt slow this quarter, can you diagnose the funnel?
CLAUDE
Pulled the trailing 90-day cohort. OUTBOUND COHORT (trailing 90 days) Meeting Set: 52 -> to SQL: 19 (37%, below 55% target) to SQO: 11 (58%, strong) -> Closed Won: 5 (45%, strong) Leaky gate: MEETING to SQL Every gate past SQL is healthy, the leak is entirely in qualification, meetings are happening but not converting into real SQLs. Key insight: this is a meeting-quality problem, not a top-of-funnel problem, so adding more outbound volume would have made the wrong number worse. One sharpener: connect the product-analytics tool so Product-channel meetings get graded on real usage instead of self-reported fit.
04 / GO FURTHER

Where an operator takes this next.

The diagnosis runs once, by hand. Here is where an operator takes it once the manual version proves out.

The framework finds the gate; the automation keeps watching it after you look away.

01
Run the diagnosis every Monday, not just when it's asked

A scheduled Claude task pulls the trailing-90 cohort from Salesforce weekly and flags the gate that moved past tolerance.

02
Catch the leak while the rep can still fix it

Wire a Slack alert the moment a meeting-to-SQL rate on a live cohort drops below target, so coaching happens mid-quarter, not in the QBR.

03
Turn Sprint Planning into a standing queue

Feed qualified accounts through Clay for buying-committee enrichment automatically, so the ranked backlog refreshes without a manual sprint kickoff.

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.