sales-methodology-implementer-skill

Implement and scale proven sales methodologies across the Epiphan sales team. Takes abstract frameworks (MEDDIC, BANT, Sandler, Challenger, SPIN, Value Selling, Gap Selling) and makes them concrete: discovery questions by component, deal scorecards scored from CRM reality rather than rep gut, training and coaching materials, and a HubSpot write-back so scores persist instead of living in disposable markdown. Keywords: `sales methodology`, `MEDDIC`, `BANT`, `Sandler`, `SPIN`, `scorecard`, `enablement rollout`

<quick_start> Mode A — "implement [methodology] across my team" (enablement package, internal):

1. Confirm inputs: methodology, deal type, average deal size, sales cycle length (defaults: Epiphan AV hardware, $10K-$150K, 1-6 months).
2. execute_sql on hubspot_deals (last 12 months, read crm://schema/hubspot_deals first) — stage durations, win rate, deal-size distribution → calibrate scoring thresholds and the 30/60/90 targets.
3. Generate the guide from reference/output-template.md. Discovery questions and call-prep scripts are spoken to prospects → check_my_copy on each (Gate B).
4. Deliver: markdown in chat + saved to the workspace (docs/ or the path Tim names); optionally create_file (Google Drive) when the team needs a shared copy.

Mode B — "score this deal using [methodology]" (one deal, evidence-based):

1. hubspot_search_deals → hubspot_get_deal (stage, amount, close date, notes, associations) → hubspot_search_contacts for the company.
2. qualify_lead per contact (Gate Q) — ATL/BTL tier + EB flag + exclusions. This is the only source for the Economic Buyer / Authority / Champion components.
3. clari_list_calls → clari_get_call_summary (last 90 days) and search_calls (company name) — evidence for Metrics, Pain, Decision Criteria/Process, Need, Timeline.
4. Score each component 0-10 with the evidence line filled from steps 1-3; overall 0-100; status QUALIFIED / PURSUE WITH CAUTION / DISQUALIFY.
5. Write back: hubspot_create_contact_note (scorecard) on the champion; hubspot_create_contact_task for the next action (confirm-first). Deal-property changes are recommended, not written. </quick_start>

<house_rules> Customer-facing skill. Before output, run the gates in order (details: ~/.claude/skills/epiphan-house-rules-skill/reference/gates.md): S: search_product_catalog (Epiphan AI) or Epiphan Knowledge search before any product claim. B: check_my_copy on every customer-facing string. Q: qualify_lead right after the candidate pull, before tiering/enrichment/drafts. D: outbound text becomes a Gmail draft, never a send. Identifiers (owner IDs, portal, emails, seats) come from reference/identifiers.md in that skill — do not restate them here. On any gate failure follow reference/self-healing.md (retry, degrade with a flag, alert Tim, halt) and append a line to ~/.claude/skill-runs/.jsonl. </house_rules>

<success_criteria>

• All framework components explained with discovery questions (Tier 1/2/3), each question brand-checked
• Deal scorecard with scoring rubric and red/green flags per component
• Mode B: stakeholder components (EB / Authority / Champion) sourced from qualify_lead; evidence lines cite HubSpot or Clari
• Mode B: scorecard persisted to HubSpot as a note; next action as a task (after confirmation)
• Training plan covering 30/60/90 day implementation, thresholds calibrated from execute_sql history
• Call prep template and manager coaching guide included
• Any product/competitor statement in example answers verified via search_product_catalog
• Outcome sidecar + run-log line written </success_criteria>

Supported Methodologies

Methodology	Best For	Cycle	Deal Size	Stakeholder component fed by qualify_lead
MEDDIC	Enterprise B2B, complex sales	3-12 mo	$50K+	Economic Buyer, Champion
BANT	SMB, transactional sales	1-4 wk	$5K-$50K	Authority
Sandler	Consultative, avoiding "free consulting"	Varies	Varies	Decision (who/how)
Challenger	Complex B2B, competitive markets	3-9 mo	$25K+	Mobilizer identification
SPIN	Complex sales, large accounts	3-12 mo	$50K+	Need-payoff owner
Value Selling	Competitive, commoditized markets	Varies	Varies	Value owner (budget holder)
Gap Selling	Change-resistant prospects	Varies	Varies	Current-state owner

Core Capabilities

• Generate methodology-specific discovery questions (Gate B on each)
• Create deal scorecards and qualification checklists
• Build rep training materials and playbooks
• Design pipeline stage gates aligned to methodology
• Develop coaching conversation guides for managers
• Score deals from live CRM/Clari evidence, identify gaps, flag disqualifications, write back to HubSpot

Stage 1: Inputs & Calibration

• Methodology, deal type, average deal size, cycle length. Ask once if missing; otherwise use Epiphan defaults.
• execute_sql on closed deals (12 months): median days per stage, win rate by stage, deal-size quartiles. Use these to set the scorecard thresholds (what "7-9 solid" means in days/dollars) and the 30/60/90 targets. If SQL is unavailable after one retry, use the generic thresholds in the template and mark the run partial.

Stage 2: Framework Guide (Mode A)

Fill reference/output-template.md: overview → per-component definition, why it matters, Tier 1/2 discovery questions (why ask / listen for / follow-up), red/green flags, 0-10 rubric → scorecard → call prep template → manager coaching guide → 30/60/90 plan → success metrics.

Gate B: every discovery question and call-prep line passes check_my_copy (they are spoken to prospects). Gate S: any Epiphan product or competitor statement used as an example answer goes through search_product_catalog first.

Stage 3: Deal Scoring (Mode B)

Component family	Evidence source (call order)
Economic Buyer / Authority / Champion / Mobilizer	hubspot_search_contacts → qualify_lead per contact: ATL tier + EB flag = EB candidate; BTL contact can never be scored as EB. Champion = highest engagement among non-excluded contacts
Metrics / Pain / Need / Value	clari_get_call_summary + search_calls (what the prospect said); HubSpot notes via hubspot_get_deal
Decision Criteria / Process / Timeline	Clari summaries, hubspot_get_deal (close date, stage history), ask_agent for the activity timeline
Budget	HubSpot amount vs. qualify_lead ICP / tier signals; stated budget in calls

Score each component 0-10 with a one-line evidence citation; no evidence = max 3 and a "Missing Information" entry. Overall 0-100 → QUALIFIED (≥70) / PURSUE WITH CAUTION (50-69) / DISQUALIFY (<50), thresholds adjusted by Stage 1 calibration. Degrade: qualify_lead down after retry → stakeholder components scored max 5 with [tier unverified], run partial; Clari empty → evidence from HubSpot notes only, flagged.

Stage 4: Write-Back (confirm-first)

• hubspot_create_contact_note on the champion / primary contact: scorecard table, status, evidence, tiers, next action. Deal link format from reference/identifiers.md (house-rules skill).
• hubspot_create_contact_task: next action with due date — propose, Tim/AE confirms, then create.
• Deal properties ([methodology]_score, [methodology]_status, qualification_date, deal_risk_level): recommend values in the output; no deal-property write tool exists in the Epiphan AI MCP as of 2026-10.
• Any prospect-facing follow-up text produced alongside the score → check_my_copy → create_draft (Gmail), never sent.

Stage 5: Delivery

• Mode A: full guide in chat + saved markdown (workspace path Tim names, default docs/methodology-<name>-<date>.md); optional Google Drive create_file for team sharing.
• Mode B: scorecard in chat + HubSpot note id + task id (or "pending confirm") + the sidecar.

Best Practices

1. Start with one methodology — don't implement several frameworks at once
2. Customize to the Epiphan motion — AV hardware, education/courts/corporate verticals, 1-6 month cycles
3. Make it easy — the scorecard lives in HubSpot notes, not a side document
4. Coach consistently — review deal scoring in every 1-on-1
5. Celebrate disqualifications — praise reps who disqualify bad deals early
6. Track over time — execute_sql score-vs-win correlation at Day 90

<reference_guides>

• reference/output-template.md — full guide / scorecard / coaching / 30-60-90 / write-back template
• Sibling skills: meddic-call-prep-auto-skill (single-call prep), challenger-sale-skill (Challenger narrative), sales-revenue-skill (MEDDIC + pipeline benchmarks), never-split-the-difference-skill (negotiation framing) </reference_guides>

<self_healing> partial = guide or scorecard produced but a stage degraded: execute_sql calibration unavailable → generic thresholds; qualify_lead down → stakeholder components capped at 5 with [tier unverified]; Clari empty → HubSpot-notes evidence only; Brand MCP down → questions prefixed [brand check skipped]; HubSpot note write failed → scorecard returned in chat with "write-back pending". error = Mode B deal cannot be resolved after one retry, or Mode A inputs missing and Tim unavailable. halt = Mode B with both qualify_lead and HubSpot contact data unavailable (a scorecard with no stakeholder truth is not a scorecard). Ladder per ~/.claude/skills/epiphan-house-rules-skill/reference/self-healing.md: retry once → degrade with a flag → alert Tim (Slack DM) → halt with written state. Append one line per run to ~/.claude/skill-runs/sales-methodology-implementer-skill.jsonl. </self_healing>

<outcome_sidecar> At the end of every run write ~/.claude/skill-analytics/last-outcome-sales-methodology-implementer-skill.json:

{"ts":"[UTC ISO8601]","skill":"sales-methodology-implementer-skill","version":"<frontmatter version>","variant":"default",
 "status":"[success|partial|error]","runtime_ms":[estimated ms from start],
 "metrics":{"mode":"A|B","methodology":"<name>","scorecards_created":[n],"discovery_questions_generated":[n],"deal_score":[n],"gate_q_contacts":[n],"hubspot_note_written":false},
 "error":null,"session_id":"[YYYY-MM-DD]"}

Use status "partial" if some stages failed but results were produced (name the stage in error). Use "error" only if no output was generated. </outcome_sidecar>