skills/sales-call-review/SKILL.md
Reviews sales calls, extract coaching insights, and score against MEDDPICC/SPIN/Challenger. Use when a rep lost a deal and you want to understand why, call went sideways and need to debrief, not sure if rep is qualifying properly, need to pull action items from a recorded call, rep talking too much and not listening, or need to give structured feedback on a call. Do NOT use for deal-level risk analysis (use /sales-deal-inspect).
npx skillsauth add sales-skills/sales sales-call-reviewInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Help the user review a sales call — score it against best practices, identify key moments, extract coaching insights, and plan follow-up actions.
If references/learnings.md exists, read it first for accumulated knowledge.
Ask the user:
What type of call was this?
What feedback are you focused on?
Do you have a transcript or recording?
Who is providing/receiving feedback?
If the user's request already provides most of this context, skip directly to the relevant step. Lead with your best-effort answer using reasonable assumptions (stated explicitly), then ask only the most critical 1-2 clarifying questions at the end — don't gate your response behind gathering complete context.
Use this scorecard as a framework — adapt the dimensions based on call type and what the user asked to focus on. Not every dimension applies to every call (e.g., objection handling may not come up in a first discovery call).
| Dimension | Score (1-5) | Evidence | Coaching note | |-----------|-------------|----------|---------------| | Opening & rapport | | Did they earn the right to continue in the first 30 seconds? | | | Discovery depth | | Did they uncover real pain, impact, and urgency? Or stayed surface-level? | | | Active listening | | Did they paraphrase, ask follow-ups, and build on what the prospect said? | | | Value articulation | | Did they connect capabilities to the prospect's specific situation? | | | Objection handling | | Did they acknowledge, clarify, and address objections effectively? | | | Next steps | | Did they secure a specific, time-bound commitment? | | | Talk-to-listen ratio | | Target: 40-60% talk for discovery, 60-70% for demos | | | Longest monologue | | Target: under 2 minutes. Long monologues lose prospects. | | | Filler words | | Excessive "um", "like", "you know", "basically" erode credibility | |
Overall score: Average of all dimensions → Rating:
MEDDPICC layer (if applicable): | Element | Addressed? | What was uncovered | What's still missing | |---------|-----------|-------------------|---------------------| | Metrics | | | | | Economic Buyer | | | | | Decision Criteria | | | | | Decision Process | | | | | Paper Process | | | | | Implicate the Pain | | | | | Champion | | | | | Competition | | | |
SPIN layer (if applicable): | Question Type | Count | Quality (1-5) | Best example | |--------------|-------|---------------|-------------| | Situation | | | | | Problem | | | | | Implication | | | | | Need-Payoff | | | |
Challenger layer (if applicable):
Identify and analyze the most important moments:
What was the single best thing the rep did on this call? Quote it if possible and explain why it worked.
What was the biggest moment they could have capitalized on but didn't? What should they have said or done instead?
For each objection that came up: | Objection | How it was handled | Better approach | Rating (1-5) | |-----------|-------------------|-----------------|---------------|
List any positive signals the prospect gave (verbal or behavioral):
List any warning signs:
Provide 3-5 specific, actionable coaching recommendations:
For each recommendation:
Extract concrete action items from the call:
List everything both sides committed to doing, with owners and deadlines.
List questions the prospect asked that weren't fully answered, plus questions the rep should have asked but didn't.
Who was mentioned on the call that needs to be brought into the process? What's the plan to reach them?
Write a follow-up email the rep can send within 24 hours:
If a follow-up call was booked, draft a proposed agenda:
For platform-specific call review guidance (Revenue.io, Seismic, ZoomInfo Chorus, Outdoo, Jiminny, Enthu.AI, NICE CXone, Sybill, Balto, Weflow), see references/platforms.md.
/sales-sybill — Sybill platform help (CRM Autofill, Deal Workspace, Ask Sybill, Magic Summaries)/sales-jiminny — Jiminny platform help (coaching-focused conversation intelligence, revenue intelligence, 8 CRM integrations)/sales-gong — Gong platform help (Smart Trackers, coaching scorecards, deal boards, transcript API)/sales-note-taker — Picking an AI note-taker (Fathom, Fireflies, Avoma, Gong, etc.) or wiring its API into CRM/warehouse for transcripts and summaries/sales-salesloft — Salesloft Conversations settings, coaching playlists, recording configuration/sales-deal-inspect — Assess overall deal health beyond just the call/sales-cadence — Optimize outbound cadences based on call learnings/sales-seismic — Seismic platform help including Meeting Intelligence and content management/sales-zoominfo — ZoomInfo platform help (Chorus conversation intelligence, Momentum deal insights)/sales-salesask — Sales Ask platform help (field + call center AI coaching, Coach Dean active feedback, home services vertical)/sales-outdoo — Outdoo platform help (AI roleplay coaching, conversation intelligence, revenue intelligence, CRM automation, mid-market Gong alternative)/sales-enthu — Enthu.AI platform help (contact center QA with auto-scoring, agent coaching, compliance monitoring, affordable Gong alternative)/sales-revenue-io — Revenue.io platform help (Salesforce-native conversation intelligence, AI scoring with 400+ criteria, Orchestrate tier)/sales-observe-ai — Observe.AI platform help (enterprise contact center QA, 100% auto scoring, Agent Copilot, VoiceAI/ChatAI agents)/sales-nice-cxone — NICE CXone platform help (built-in QM with evaluation forms, screen + voice recording, Interaction Analytics)/sales-balto — Balto platform help (real-time AI coaching for contact centers, automated QA, compliance monitoring)/sales-salesken — Salesken platform help (real-time in-call coaching, QA automation, revenue intelligence, multilingual transcription)/sales-clari-copilot — Clari Copilot platform help (real-time battlecards, live coaching, AI call summaries, deal intelligence)/sales-weflow — Weflow platform help (Salesforce-native Revenue AI with activity capture, conversation intelligence, deal forecasting)/sales-do — Not sure which skill to use? The router matches any sales objective to the right skill. Install: npx skills add sales-skills/sales --skill sales-doDon't score a call without a transcript or detailed description. Claude will confidently generate scores from thin air if given only "it was a discovery call." Push back and ask for specifics — what was said, how it ended, what objections came up.
Don't focus only on negatives. Claude tends to generate long lists of improvements. Always lead with the best moment and what went well before coaching points. Reps shut down if feedback feels like a list of failures.
Don't apply MEDDPICC scoring to a first discovery call. MEDDPICC elements like Paper Process and Competition are rarely uncovered in a first meeting. Score only the elements that were reasonable to address at this deal stage.
Don't ignore talk-to-listen ratio. This is one of the most objective and actionable metrics. If the user provides a transcript, estimate the ratio. If they describe the call, ask about it directly.
Don't treat the scorecard as a final verdict. A single call is a snapshot. Frame scores as "on this call" not "this rep is a 2/5 at discovery." Context matters — a rough call with a hostile prospect is different from a rough call with a warm lead.
Self-improving: If you discover something not covered here, append it to references/learnings.md with today's date.
User says: "Review this discovery call — I talked for 70% of it, asked about their process, showed a quick demo, and ended with 'I'll send you info and we can reconnect next week.'" Skill does:
User says: "Score my rep's call using MEDDPICC — they confirmed pain and got metrics, but didn't ask about decision process or economic buyer." Skill does:
Solution: Have the rep describe what happened in detail — key moments, objections raised, how the call ended, and what was agreed. The skill can still provide a useful review from a narrative description, though transcript-based reviews are more precise.
Solution: Use the "best moment" analysis first to lead with what went well. Frame coaching points as "opportunities" not "mistakes." Focus on 1-2 highest-impact areas rather than listing every issue. The self-review mode (Step 1, option B) helps reps build self-awareness before receiving external feedback.
Solution: Ground every score in specific evidence from the call. Use the methodology layers (MEDDPICC, SPIN, Challenger) for more objective, criteria-based assessment. Compare scores across multiple calls to calibrate — a single call review is a snapshot, not a verdict.
tools
Wizlogo (wizlogo.com) platform help — a budget online logo maker (template/style-variation, marketed as "AI") plus a hub of FREE branding tools (business-name, blog-name and slogan generators, business-card maker, invoice generator, color converter, domain search). The pricing traps: the FREE logo is PERSONAL-USE-ONLY; the two cheap paid tiers are RASTER PNG/JPG only — Single (~€39.99 one-time) and Unlimited (~€3.99 per WEEK, recurring) — and VECTOR (SVG/PDF/EPS) is gated to the ~€299.99 Enterprise tier, which also bundles human designer edits and a social kit. Transparent PNG is on all paid plans. Use when making a Wizlogo logo, understanding free-vs-paid or personal-vs-commercial use, which tier unlocks vector/SVG for print, the weekly-subscription billing trap, its free name/slogan generators, or whether it has an API (UI-only — no public API, webhooks, Zapier or MCP). Do NOT use to just generate the business name (use /sales-namelix) or to compare/validate branding tools (use /sales-idea-validation).
tools
VistaPrint platform help (vistaprint.com, a Cimpress company) — the small-business design + print + digital-marketing platform: a free AI Logomaker (4 generations, 60 more after free sign-up) exporting SVG/PNG/PDF at 4000x4000 with no watermark, a free Brand Kit, business cards/flyers/signage/apparel/promo print, and a website builder. THE RIGHTS TRAP: VistaPrint states NO intellectual-property rights transfer on an AI-generated logo — you get usage rights but CANNOT register it for trademark or copyright; only its human designer service transfers full IP. Use when making a VistaPrint logo, asking if you own or can trademark it, running out of AI logo credits, printed colors not matching the screen, bleed/DPI/font file-prep rejections, or asking whether VistaPrint has an API (the consumer site does not — automation runs through the parent Cimpress Open partner-fulfilment API). Do NOT use for Vista Social scheduling (use /sales-vistasocial) or comparing logo tools market-wide (use /sales-idea-validation).
tools
Turbologo (turbologo.com) platform help — a budget AI/DIY logo maker: enter a business name + industry, pick icons and colors, and it proposes logo concepts you refine in an in-browser editor, then pay a one-time fee to download (designing is free, previews are watermarked, downloading is the paywall). Vector SVG/PDF is gated to the mid tier and up; the top tier adds a brand kit (business cards, letterheads, email signatures, social assets). Use when generating a logo in Turbologo, choosing which download tier to buy, vector SVG vs raster PNG, removing the free watermark, the time-limited edit-after-purchase window, pay-to-download pricing questions, whether an AI logo is yours to trademark, or whether Turbologo has an API to bulk-generate logos (it is UI-only — no public API, webhooks, Zapier, or MCP). Do NOT use to generate the business name (use /sales-namelix), compare or validate branding tools across the market (use /sales-idea-validation), or build wider marketing creative (use /sales-canva).
tools
Online Logo Maker (onlinelogomaker.com) platform help — a long-standing free/freemium DIY logo maker: build the mark yourself from icons, shapes, text, and fonts — MANUAL/template-based, NOT enter-a-name-get-AI-concepts. The free pack downloads a LOW-RES 300px PNG with a background; vector SVG, transparent PNG, and 2000px high-res are gated to a one-time lifetime Premium pack (not a subscription). The free tier's commercial-use rights are disputed by reviewers — clean ownership effectively needs Premium, and a shared-icon mark can be non-distinctive. Use for building/editing a logo here, free download vs Premium, vector SVG or transparent PNG, one-time pricing, commercial-use/trademark terms, near-namesake confusion (NOT LogoMaker.com / LogoMakr / Logomakerr.ai), or whether it has an API (UI-only — no API, webhooks, Zapier, MCP). Do NOT use to generate the business name (use /sales-namelix), compare branding tools across the market (use /sales-idea-validation), or build wider creative (use /sales-canva).