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.
Install it in one line, or paste it in.
~/.claude/skills/ and runs automatically when it is relevant.Connect your context. Set it to your motion.
cohort dates, stages, scores, contacts, and owners, live.
real product engagement for the Product channel.
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 this | What it is | Default / Example |
|---|---|---|
| CRM connector | Where cohort dates, stages, scores, contacts live | Your CRM |
| Product-analytics connector | Where product-usage events live | Your product-analytics tool |
| ICP-qualification connector | Independent fit qualifier (optional) | Your ICP tool |
| Community/intent connector | Community + buying-committee signals (optional) | Your community tool |
| Enrichment connector | Verified contact data on top-tier accounts (optional) | Your enrichment tool |
| Channels | Your funnel entry channels | Inbound / Outbound / Product |
| Cohort anchor field per channel | The date field that marks funnel entry | A "qualified date" field per channel |
| Qualification stage | The CRM stage that equals SQL | Your "Qualification" stage |
| Discovery stage | The CRM stage that equals SQO | Your "Discovery" stage |
| New-business filter | The opportunity-type filter that excludes renewals | Opportunity type = New Business |
| Fit-score field | The model score you are auditing | Your scoring model's fit score |
| Score floor value | The value your engagement score collapses to when usage is sparse | Your model's floor score |
| Conversion benchmarks | Per-gate target conversion rates | Your trailing baseline |
| Average deal size | Used to dollarize leverage points | Your current ASP |
Everything the skill does, in full.
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.
- 1The 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:
- Cohort created (a qualified-account date fires) to Meeting booked.
- Meeting booked to Meeting completed (show rate).
- Meeting completed to SQL (your Qualification stage).
- SQL to SQO (your Discovery stage).
- SQO to Closed Won.
- 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.
- 2Mode 1: Funnel Diagnosis
Default mode is conversational. Work interactively. Do not dump a full report unprompted.
- Anchor on a channel and a cohort window. Ask what they want to diagnose.
- Pull live data from the CRM. Always run queries. Do not analyze from memory.
- 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.
- Diagnose with the pattern framework. Each leaky gate has two to four known failure modes. Identify which fits.
- Recommend a specific action. Named: a person, a list of accounts, a process change, a target.
- Offer the written write-up only if asked.
- 3Mode 2: Sprint Planning
Converts the funnel from a diagnostic into an execution backlog.
- Define the cohort. Default: all accounts whose qualified-account date falls in the target window.
- 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).
- Compute composite scores. Tier into Hot, Warm, Watch, Disqualify.
- Build the spreadsheet: Sprint Ranking, Top Buying Committee, Scoring Rubric, Cohort Summary, Scoring Gaps.
- 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.
- 4Mode 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.
- 5Action 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.
- 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.
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
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.
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.
A scheduled Claude task pulls the trailing-90 cohort from Salesforce weekly and flags the gate that moved past tolerance.
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.
Feed qualified accounts through Clay for buying-committee enrichment automatically, so the ranked backlog refreshes without a manual sprint kickoff.
One skill is the on-ramp.
A single skill does one job. Chained into a playbook, or run as a full build, it becomes a system. Here is where this one plugs in.