evals/evals.json
{
"skill_name": "meddpicc-scorecard",
"evals": [
{
"id": 1,
"prompt": "Score this deal for me. Halberd Freight, 240-seat rollout of our TMS, $195K ACV, six people involved on their side, they have formal procurement plus a SOC 2 review to get through, we're at proposal. Metrics: about $600K a year saved on empty miles, I built the model off their volumes. EB is the COO, my contact says he's fully behind it. Criteria: their RFP lists 14 requirements. Process: they told me they want it signed by end of Q3. Paper: procurement there is a rubber stamp, my contact has done it before. Pain: their dispatchers hate the tool they've got. Champion: the ops manager loves us, we talk every week. Competition: nobody else is in it. I've got it at 19 out of 24, does that look right?",
"expected_output": "An eight-element scorecard in which every claim lacking a buyer-side artefact is knocked down to 1, the self-scored 19/24 is rejected, the Champion gate caps the verdict at Under-qualified, each knocked-down score appears under Challenged with its reason, and the top next action is a buyer-visible test.",
"files": [],
"expectations": [
"Scores all eight elements, since the deal is $195K with six stakeholders and a formal procurement and security path.",
"Scores Metrics 1, on the ground that the savings figure is a rep-built model the buyer never confirmed.",
"Scores Economic Buyer 1 or 0, treating 'my contact says he's fully behind it' as hearsay rather than the Access rung, since no introduction has been committed.",
"Scores Decision Criteria 1, on the ground that an RFP requirement sheet is not buyer-confirmed decision criteria.",
"Scores Decision Process 1, on the ground that the end-of-Q3 date is not a buyer-owned decision process.",
"Scores Paper Process 1, on the ground that 'procurement is a rubber stamp' names no mapped path, no owners, and no timelines.",
"Scores Implicate the Pain no higher than 2, because the dispatcher complaint carries no costed consequence.",
"Scores Champion 1, identifying the ops manager as a coach who has not acted.",
"Scores Competition 1, noting that do-nothing or inertia has not been examined.",
"Rejects the self-scored 19/24 and reports a materially lower total.",
"States that the Champion gate caps the verdict at Under-qualified regardless of the total.",
"Includes a Challenged section listing each knocked-down score together with the missing artefact as the reason.",
"The single top next action is an ask the buyer either performs or refuses, not rep-side work such as updating the CRM or building a deck."
]
},
{
"id": 2,
"prompt": "Quick one. $11K a year, it's a solo founder, he's the only person involved, no procurement, no legal team, no security review. He told me on Friday he'd sign this month. Run MEDDPICC on it and tell me if it's real.",
"expected_output": "A right-sized scorecard covering only the letters that survive for a deal this size, with Economic Buyer folded into Champion, a shrunken denominator, and no next action attached to any dropped element.",
"files": [],
"expectations": [
"Right-sizes the element list before scoring instead of scoring all eight elements.",
"Scores only Metrics, Implicate the Pain, and Champion, adding Decision Process solely if a multi-step approval is reported.",
"States that Economic Buyer and Champion collapse into the same person on this deal and scores them as one letter rather than dropping Economic Buyer outright.",
"Explicitly drops Paper Process, Decision Criteria, and Competition from the scored list.",
"Uses a denominator matching the reduced element list rather than 24.",
"Attaches no next action to any element that was right-sized out.",
"States that scoring all eight letters on a deal this size would be process theatre or over-engineering.",
"Treats the 'he'd sign this month' date as rep-reported rather than buyer-owned evidence of a decision process.",
"Confirms the element list with the user before scoring, or states the right-sizing decision and its reason explicitly in the output.",
"States the definitions used for the second C and for Pain, and that weighting is equal by default.",
"Presents disqualification or a low band as a legitimate outcome rather than forcing a path forward."
]
},
{
"id": 3,
"prompt": "I run a 14-rep team at Corvid Analytics, $2.1M quarterly quota, our forecast accuracy is sitting around 55%. I'm rolling deal scoring out across the team before next quarter. Build me the rubric the reps will score their opportunities on, and the slide that tells my VP what we should expect to get out of it.",
"expected_output": "An evidence-based 0-3 rubric with equal weighting stated, plus an expectations slide that quarantines unreplicated vendor outcome figures, labels any vendor number it does quote, and points at the one measure that can falsify the rubric.",
"files": [],
"expectations": [
"Defines the scoring ladder as 0 unknown, 1 assumed, 2 stated, 3 validated, with 2 meaning the buyer said it and 3 meaning the buyer demonstrated it.",
"States that the axis is evidence rather than presence: a letter scores on what the buyer did, not on what the rep can name.",
"States that weighting is equal by default and that per-letter weighting is practice-dependent rather than canonical.",
"Does not present '18% higher win rates', '24% larger deals', or '26% shorter sales cycles' as expected outcomes.",
"Does not promise a forecast-accuracy figure such as 95%.",
"Does not cite 'increase deal size by 62%' or 'double your win rate' as an expected result.",
"Does not cite '40% higher close rates', '78% of accounts are single-threaded', or a 2.4x or 3.1x multi-threading multiplier.",
"Labels any vendor-published figure it does quote as a vendor claim rather than an independent measurement.",
"Attributes the 60% lost-to-inertia figure to the methodology publisher's own claim if it cites it at all.",
"Describes any Ebsta or Pavilion benchmark figure it cites as a relative lift rather than a point difference.",
"Warns that scorecards getting filled while forecast accuracy stays flat means the evidence definitions are too loose.",
"Names band versus realized outcome as the measure that can falsify the rubric, and notes it only pays out after a full quarter of closed deals.",
"Distinguishes rules published by the methodology from rules this scorecard sets itself."
]
},
{
"id": 4,
"prompt": "Deal's already scored, I just need to know what to chase. Wexler Industrial, $120K ACV, seven stakeholders, they have procurement but we're nowhere near a yes yet, mid-cycle. Scores: Metrics 1, Economic Buyer 1 (I know the CFO's name, no intro on the table), Decision Criteria 1, Decision Process 2, Paper Process 1, Pain 2, Champion 3 because he's presented us internally twice and sent me their scoring sheet, Competition 1. Forecast call is Friday. One thing though: I've already burned a lot of goodwill with my champion getting the security questionnaire answered, so I'm not asking him for anything that costs him internally. What do I go do?",
"expected_output": "A ranked ask list in which the effort ceiling deletes the Economic Buyer meeting and the procurement introduction outright, the Friday checkpoint demotes the slow asks, and the recommendation is a near-zero buyer-visible ask.",
"files": [],
"expectations": [
"Deletes the Economic Buyer meeting from the ask menu rather than ranking it last, because the stated effort ceiling forbids spending the champion's internal credit.",
"Deletes the procurement introduction on the same ground.",
"States which asks were deleted and the constraint that removed them.",
"Does not recommend the Economic Buyer meeting as the next action despite Economic Buyer being among the weakest-evidenced elements.",
"Recommends a near-zero ask that fits inside a conversation already scheduled.",
"Ranks the remaining asks by verdict change per unit of effort rather than by element importance.",
"States effort as an order of magnitude such as near-zero, an hour, or a week, never as a currency amount.",
"Notes that the Friday checkpoint demotes the paper-process map because it cannot land in time.",
"Attaches no next action to Champion, which scores 3.",
"Lists the starved high-value asks under a Deferred heading together with the condition that would promote them.",
"Applies no gate cap, since Champion is above 1 and Economic Buyer is above 0.",
"Keeps Economic Buyer at 1 rather than raising it on the strength of the champion's internal presentations, because no introduction has been committed."
]
},
{
"id": 5,
"prompt": "Ferrous Labs, $310K over three years, nine people involved, legal and infosec both in the loop, we're in technical validation. Honestly nobody on my side knows who actually signs. My champion is a platform director, he's presented us internally twice and sent me their scoring sheet. They emailed me that the March outage cost them $1.4M. Their target is cutting incident MTTR by 40%, that's their own number out of their board deck. Criteria are in the scoring sheet they sent. They walked me through the committee steps on a recorded call, but there's nothing written down between us. Nobody has mentioned procurement at all. They named two rivals and said doing nothing is genuinely on the table. Score it.",
"expected_output": "An eight-element scorecard with several genuine 3s, an Economic Buyer at 0 that caps the verdict at Under-qualified regardless of the total, and the Economic Buyer meeting promoted ahead of cheaper asks on the stated gate condition.",
"files": [],
"expectations": [
"Scores all eight elements, since the deal is $310K with nine stakeholders and legal and infosec involved.",
"Scores Economic Buyer 0 and states the verdict is capped at Under-qualified regardless of the total.",
"Does not report the band as Forecast-ready or Real but gapped despite a raw percentage in the 55-79% range.",
"Scores Champion 3, citing the internal presentations and the shared scoring sheet as observable acts.",
"Scores Metrics 3, citing the buyer-supplied MTTR target from their own board deck.",
"Scores Implicate the Pain 3, citing the emailed $1.4M outage cost as a costed consequence the buyer stated.",
"Scores Decision Process 2 rather than 3, because the steps were described but no plan was co-edited with the buyer.",
"Scores Competition 2 rather than 3, because alternatives were named but no win plan positions against do-nothing.",
"Scores Paper Process 1 or 0, since procurement has never been discussed.",
"Names the Economic Buyer meeting as the first action, overriding the default efficiency order.",
"States that the promotion condition, the gate capping the verdict, put the Economic Buyer meeting ahead of cheaper asks, rather than its value alone.",
"Phrases the top action as an ask the champion either performs or refuses.",
"Notes that a champion unwilling or unable to make the introduction would itself call the Champion score into question."
]
},
{
"id": 6,
"prompt": "I sell residential solar. Couple in their fifties, $61K system, one home visit done. The wife is keen, she's the one who called us after a neighbour's referral. The husband keeps saying the payback maths doesn't work. They'd need financing and haven't applied anywhere. They told me their power bill is $480 a month and climbing. Is this one real? Run the qualification framework over it.",
"expected_output": "A four-element household scorecard, labelled as a structural analogy, keeping the financing path in the scored set and promoting it, with the skeptical spouse identified as the money-controlling role.",
"files": [],
"expectations": [
"States that the consumer mapping is a structural analogy rather than established practice for this methodology.",
"Right-sizes to roughly four elements rather than scoring all eight on a household purchase.",
"Keeps Paper Process in the scored set, read as the financing path.",
"States that Paper Process inverts the usual drop order here and is promoted by default rather than dropped first.",
"Maps the skeptical husband to the money-controlling, veto-holding role rather than treating the enthusiastic wife as sufficient.",
"Maps Champion to the household member who sells the other when the salesperson is not present.",
"Scores the financing element 1 or 0, since no application has been started.",
"Scores Metrics no higher than 2: the $480 bill is stated by the buyer but the lender's figure is unconfirmed.",
"Names waiting or doing nothing as the live competing alternative even if Competition is right-sized out of the scored list.",
"Recommends a buyer-visible test such as getting both spouses into the same conversation, or a financing application actually submitted by a stated date.",
"Keeps the evidence ladder, the anti-inflation rule, and the gates identical to the business-to-business version, changing only the cast.",
"Does not present a full eight-element scorecard as the more rigorous option for this deal."
]
},
{
"id": 7,
"prompt": "Our org runs this framework but our second C is Compelling Event, not Competition, and we weight Economic Buyer and Champion double. Score this one under our model. Nimbus Freight Systems, $140K ACV, five stakeholders, procurement is definitely involved, we're at business case. My champion is the VP Ops, she's booked me into the CFO's calendar for next Thursday. Metrics: I put $900K of avoided overtime in front of her and she said that sounds about right. Pain: she said the overtime problem cost them a customer last year but wouldn't put a figure on it. Criteria: she walked me through them on a call, nothing captured. Decision process: same, described verbally, nothing written between us. Paper: she says security review runs six weeks and legal another two, she's watched it happen twice. Compelling event: their WMS contract expires in March and they can't renew it.",
"expected_output": "A scorecard run under the team's own definitions and weighting, with the do-nothing probe deleted from the ask menu, a mostly-2 profile reflecting verbal-only evidence, and no gate cap.",
"files": [],
"expectations": [
"Uses the team's Compelling Event as the second C instead of Competition and states that substitution in the output.",
"Applies the team's double weighting for Economic Buyer and Champion instead of equal weighting, and states the weighting used.",
"Deletes the do-nothing probe from the ask menu, since the team standard replaced Competition.",
"Notes the mapping between the team's definitions and this scorecard's own defaults.",
"Scores Economic Buyer 2, the Access rung, because the introduction is committed but the meeting is not yet held.",
"Scores Champion 3, citing the booked CFO introduction as the act.",
"Scores Metrics 2 rather than 3, because the buyer verbally agreed once to a figure the rep proposed.",
"Scores Implicate the Pain 2 rather than 3, because the problem is named but uncosted.",
"Scores Decision Criteria 2 rather than 3, because the criteria were described verbally and never captured.",
"Scores Decision Process 2 rather than 3, because no plan was co-edited with the buyer.",
"Scores Paper Process 2 rather than 3, because the path is sketched but not mapped with named owners and timelines.",
"Applies no gate cap, since Champion is 3 and Economic Buyer is 2.",
"Does not add Competition as a ninth scored letter alongside the team's Compelling Event."
]
},
{
"id": 8,
"prompt": "Already scored mine, came out 15 out of 24. Metrics 2, Economic Buyer 1, Decision Criteria 2, Decision Process 3, Paper Process 1, Pain 3, Champion 2, Competition 1. Everything I know about this account came from one guy, Dmitri, their engineering manager, great bloke, tells me everything. Give me my action list for the week.",
"expected_output": "A ratio-ordered action list with one singled-out top ask, every action phrased as a buyer-visible test, single-threading flagged, and the Decision Process 3 re-challenged against the Economic Buyer 1.",
"files": [],
"expectations": [
"Flags single-threading, since one contact supplies the evidence for every element.",
"Re-challenges the Decision Process 3 against the Economic Buyer 1 as a likely inflation artifact rather than accepting both.",
"Attaches a next action to every element scoring below 3.",
"Attaches no next action to the elements scoring 3.",
"Phrases every action as an ask the buyer performs or refuses, never as rep-side work such as updating the CRM or drafting an internal note.",
"Orders the actions by verdict change per unit of effort rather than by element order or by importance.",
"Singles out one top action instead of handing over a flat list.",
"States effort per action as an order of magnitude, never as a currency amount.",
"Names the starved high-value asks under a Deferred heading with the condition that would promote them.",
"Notes that inertia is unexamined and that Competition cannot exceed 2 without a do-nothing analysis.",
"Reports a band of Real but gapped or lower, never Forecast-ready.",
"Names a re-score trigger: the event or date for the next scoring pass."
]
},
{
"id": 9,
"prompt": "Can you MEDDPICC my deal for me? It's an enterprise one, been dragging since April.",
"expected_output": "A single opening question, not a scorecard: the interview opens on deal-frame questions one at a time with multiple-choice options, and nothing is scored before the element list is confirmed.",
"files": [],
"expectations": [
"Asks questions rather than producing a scored scorecard from the information given.",
"Asks one question per message rather than presenting a numbered list of questions in a single message.",
"Offers multiple-choice answers wherever the question admits them.",
"Asks deal-frame questions, such as deal type, size and stakeholder count, stage, and who set the expected close date, before any per-element evidence question.",
"Asks whether procurement, security review, or legal redlines are involved on the buyer's side.",
"Asks whether the user's team mandates its own definitions, scale, or weighting, and commits to using theirs if so.",
"Asks by what date the scorecard must change something.",
"Asks whether the user wants a one-off read or a record that compounds across reviews.",
"Asks what the effort ceiling is: how many buyer asks can be made and how much champion credit can be spent.",
"Confirms the element list with the user before scoring anything.",
"Invents no scores, artefacts, or buyer statements the user has not supplied.",
"Asks, per element, for the buyer-side artefact or event behind the claim rather than for the claim alone."
]
},
{
"id": 10,
"prompt": "Is this one real? $47K, three people on their side, mid-stage. I bumped into their finance director at a conference last month and she told me the budget's approved for this. My main contact told me over drinks that they're definitely going with us. I've got nothing in writing from them at all, every note is mine. The close date in the CRM is one I put in, next month.",
"expected_output": "A right-sized scorecard where unrecorded verbal recall scores as assumption, the Champion gate caps the verdict, the band is stated in forecast terms, and re-qualifying or disqualifying is offered as a valid outcome.",
"files": [],
"expectations": [
"Applies the rule that evidence must exist in writing or on a recorded call, so hallway and over-drinks recollections do not count as buyer-side evidence.",
"Scores Economic Buyer no higher than 1 despite the reported 'budget is approved' conversation.",
"Scores Champion no higher than 1, since the contact's enthusiasm comes with no observable act.",
"Applies the Champion gate and caps the verdict at Under-qualified or lower.",
"Treats the close date as rep-set rather than buyer-owned and scores Decision Process accordingly.",
"Right-sizes this deal by dropping Paper Process, since no procurement is reported.",
"States that re-qualifying from discovery or disqualifying is a legitimate outcome.",
"Names exactly one element as the biggest gap.",
"Gives that gap a buyer-visible test whose refusal is itself informative.",
"States the band in forecast terms rather than as a grade or a mark out of ten.",
"Records the compelling event as honestly absent rather than manufacturing one.",
"Uses a denominator matching the reduced element list rather than 24."
]
}
],
"trigger_queries": [
{ "query": "Score this opportunity against MEDDPICC", "should_trigger": true },
{ "query": "Run a MEDDIC check on my biggest deal", "should_trigger": true },
{ "query": "is this deal real?", "should_trigger": true },
{ "query": "I need to qualify this opportunity before the forecast call", "should_trigger": true },
{ "query": "BANT this lead for me", "should_trigger": true },
{ "query": "Help me prep my biggest deal for Thursday's deal review", "should_trigger": true },
{ "query": "how qualified is this deal actually", "should_trigger": true },
{ "query": "my manager says my pipeline is inflated, start with this deal", "should_trigger": true },
{ "query": "Grade my opportunity on the eight elements", "should_trigger": true },
{ "query": "What's missing from this deal before I commit it to the forecast?", "should_trigger": true },
{ "query": "Should I forecast this one as commit?", "should_trigger": true },
{ "query": "I can't tell if this deal is real or just theatre", "should_trigger": true },
{ "query": "walk me through meddpicc for this account", "should_trigger": true },
{ "query": "qualification scorecard for a $200k enterprise deal", "should_trigger": true },
{ "query": "Do I have enough evidence to call this deal qualified?", "should_trigger": true },
{ "query": "we have no idea who signs the contract, is that a dealbreaker", "should_trigger": true },
{ "query": "Rate my deal 0 to 3 on each qualification element", "should_trigger": true },
{ "query": "sanity check this opportunity for me", "should_trigger": true },
{ "query": "my rep swears this closes this month and I don't believe him", "should_trigger": true },
{ "query": "How do I know if this deal is qualified or just busy work?", "should_trigger": true },
{ "query": "Give me a deal qualification scorecard I can fill in", "should_trigger": true },
{ "query": "What should I go validate next on this deal?", "should_trigger": true },
{ "query": "Which gap should I chase first on this opportunity?", "should_trigger": true },
{ "query": "Is my opportunity forecast-ready?", "should_trigger": true },
{ "query": "Score this: $180K ACV, proposal stage, here's what I know", "should_trigger": true },
{ "query": "I need a qualification read on this before the QBR", "should_trigger": true },
{ "query": "we keep losing to no-decision, check this one for me", "should_trigger": true },
{ "query": "Should this deal stay in my pipeline?", "should_trigger": true },
{ "query": "my forecast keeps slipping, start with this deal", "should_trigger": true },
{ "query": "assess this opportunity's health", "should_trigger": true },
{ "query": "The customer says they love us but nothing is moving. What's actually missing?", "should_trigger": true },
{ "query": "Should I disqualify this?", "should_trigger": true },
{ "query": "run the qualification framework on a solar deal, couple, $60k system", "should_trigger": true },
{ "query": "high ticket B2C qualification, kitchen remodel, are these real buyers", "should_trigger": true },
{ "query": "score this deal where I've never met the budget holder", "should_trigger": true },
{ "query": "how do I score decision process when they won't give me a plan", "should_trigger": true },
{ "query": "grade this deal on evidence, not on my gut", "should_trigger": true },
{ "query": "what's my biggest gap on this opportunity and what do I do about it", "should_trigger": true },
{ "query": "deal inspection on one opportunity please", "should_trigger": true },
{ "query": "pressure test this deal for me", "should_trigger": true },
{ "query": "I've got one ask left with my champion, what do I spend it on?", "should_trigger": true },
{ "query": "Qualify out or push? $9K, single founder, says he'll sign this month", "should_trigger": true },
{ "query": "Build me a rubric my reps score their open deals against", "should_trigger": true },
{ "query": "re-score my deal after this week's meetings", "should_trigger": true },
{ "query": "how many of the eight letters do I actually need on a small deal?", "should_trigger": true },
{ "query": "their RFP lists all their requirements, does that count as decision criteria?", "should_trigger": true },
{ "query": "I want to know whether to commit this deal or push it a quarter", "should_trigger": true },
{ "query": "what evidence do I need before I can call this qualified?", "should_trigger": true },
{ "query": "opportunity scoring for enterprise deals", "should_trigger": true },
{ "query": "my boss wants a qualification score on every deal over 100k, start with this one", "should_trigger": true },
{ "query": "we're at verbal yes and I have no idea how their legal process works", "should_trigger": true },
{ "query": "tell me honestly whether this deal survives scrutiny", "should_trigger": true },
{ "query": "help me get rid of the happy ears on this account", "should_trigger": true },
{ "query": "score the deal: four stakeholders, $75K, no procurement", "should_trigger": true },
{ "query": "which element is weakest on this deal?", "should_trigger": true },
{ "query": "Build me a stakeholder map for this account", "should_trigger": false },
{ "query": "Who are the blockers in this buying committee?", "should_trigger": false },
{ "query": "Map my buyer's org chart and tell me who influences whom", "should_trigger": false },
{ "query": "Scan these call notes for red flags", "should_trigger": false },
{ "query": "Read my CRM activity log and tell me what looks worrying", "should_trigger": false },
{ "query": "Grade this call transcript", "should_trigger": false },
{ "query": "Score my rep's discovery call against a rubric", "should_trigger": false },
{ "query": "Build the ROI model for this deal", "should_trigger": false },
{ "query": "What's the payback period if they buy at $120K?", "should_trigger": false },
{ "query": "Write the business case my champion can forward internally", "should_trigger": false },
{ "query": "Give me discovery questions for a first call with a CFO", "should_trigger": false },
{ "query": "What should I ask on the qualification call?", "should_trigger": false },
{ "query": "Write the recap email after today's meeting", "should_trigger": false },
{ "query": "Turn these notes into next steps with owners and dates", "should_trigger": false },
{ "query": "How do I handle 'we don't have budget this year'?", "should_trigger": false },
{ "query": "They say we're too expensive, what do I say back?", "should_trigger": false },
{ "query": "Plan my concessions before the pricing call", "should_trigger": false },
{ "query": "What can I give away and what do I ask for in return?", "should_trigger": false },
{ "query": "Write my cold call opener for a VP of Ops", "should_trigger": false },
{ "query": "My outbound emails are landing in spam", "should_trigger": false },
{ "query": "Test these three subject lines for me", "should_trigger": false },
{ "query": "Design the outbound cadence, how many touches and when", "should_trigger": false },
{ "query": "Find me a personalization angle from this funding round", "should_trigger": false },
{ "query": "Define our ICP from last year's closed-won data", "should_trigger": false },
{ "query": "Segment our accounts by fit score", "should_trigger": false },
{ "query": "Where should the tier cutoffs sit and how many accounts per rep?", "should_trigger": false },
{ "query": "How much pipeline do I need to hit $4M?", "should_trigger": false },
{ "query": "Set quotas for next year across the team", "should_trigger": false },
{ "query": "Design the comp plan for our new AE seats", "should_trigger": false },
{ "query": "Should we go product-led or sales-led?", "should_trigger": false },
{ "query": "Restructure my sales org, pods or assembly line?", "should_trigger": false },
{ "query": "Estimate TAM and SAM for mid-market logistics", "should_trigger": false },
{ "query": "Build the interview loop for an AE hire", "should_trigger": false },
{ "query": "How do I get promoted from SDR to AE?", "should_trigger": false },
{ "query": "Which sales newsletters and podcasts should I follow?", "should_trigger": false },
{ "query": "Route this request to the right sales skill", "should_trigger": false },
{ "query": "Score this candidate against our hiring scorecard", "should_trigger": false },
{ "query": "Build a lead scoring model for inbound MQLs", "should_trigger": false },
{ "query": "Score the vendors bidding on the RFP we're issuing", "should_trigger": false },
{ "query": "Design our CRM's opportunity stage definitions and exit criteria", "should_trigger": false },
{ "query": "Roll up my team's forecast for the quarter across 340 open opportunities", "should_trigger": false },
{ "query": "Write a customer health score for our post-sale accounts", "should_trigger": false },
{ "query": "Calculate churn risk on this renewal", "should_trigger": false },
{ "query": "Audit our pipeline hygiene: stale deals, missing close dates", "should_trigger": false },
{ "query": "What was our win rate by segment last quarter?", "should_trigger": false },
{ "query": "Build a QBR deck for our biggest customer", "should_trigger": false },
{ "query": "Review this MSA for risky clauses", "should_trigger": false },
{ "query": "Add MEDDPICC fields to our CRM opportunity object and set the picklist values", "should_trigger": false },
{ "query": "Compare MEDDIC, SPIN and Challenger and tell me which methodology to adopt", "should_trigger": false },
{ "query": "Build an opportunity solution tree for our onboarding problem", "should_trigger": false },
{ "query": "Create a marketing scorecard for our campaign channels", "should_trigger": false },
{ "query": "Find me the best deal on a laptop under $1,200", "should_trigger": false },
{ "query": "Rate these twelve partner applications against our program criteria", "should_trigger": false },
{ "query": "Write the security questionnaire response for their infosec review", "should_trigger": false },
{ "query": "Coach my rep on objection handling from last week's calls", "should_trigger": false }
]
}
SKILL.md
---
name: meddpicc-scorecard
description: Scores one deal against the eight MEDDPICC elements on an evidence ladder - unknown, assumed, stated, validated - with a scorecard, a verdict band, and the single next action that closes the biggest gap. Full MEDDPICC for complex deals, a lightweight subset for small ones, covering B2B and considered high-ticket B2C. Use whenever the user mentions MEDDPICC, MEDDIC, BANT, deal qualification, "score this opportunity", "is this deal real", or deal-review prep, even without naming a framework. Do NOT use for stakeholder mapping (mbfinotti/sales-skills@deal-champion-mapping), note-level risk scanning (mbfinotti/sales-skills@deal-red-flags), or transcript grading (mbfinotti/sales-skills@sales-call-review).
license: MIT
metadata:
author: Maya-Beth Finotti
version: "1.3.9"
---
# MEDDPICC Scorecard
Score one described deal against the eight MEDDPICC elements and say plainly what is missing, what is assumed rather than validated, and what the rep should go validate next.
The largest field study available (Ebsta x Pavilion 2023 benchmark, 3.2M opportunities across 364 companies) found:
- 61% of methodology-adopting companies chose MEDDPICC.
- Only 15% of opportunities were fully qualified.
- Only 5% of companies scored confidence per criterion.
The acronym is everywhere; the discipline is rare - that gap is "MEDDPICC theatre", and this scorecard is built to prevent it.
The whole scorecard runs on one axis: **evidence, not presence**. A letter is never scored on whether the rep can name something; it is scored on whether the buyer has done something observable to prove it.
The skill scores one deal at a time, gathers no new information itself, and grades only what the user reports, then tells them what to go validate next.
Two confidence distinctions run throughout and belong in the output:
- Whether a rule is published by the methodology or set by this skill.
- Whether a figure is an independent measurement or a **vendor claim**, a vendor's own correlational number never independently replicated.
## Interview
Ask questions under these constraints:
- One question per message.
- Multiple-choice answers wherever possible.
- Skip anything the deal description already answers.
Two rounds:
Deal frame (right-sizing inputs, always first):
1. B2B, or considered high-ticket B2C (real estate, automotive, solar, remodel, private education, financial advisory - not impulse retail)?
2. Rough deal size (contract value or ACV), and how many people on the buying side are involved?
3. What stage is the deal at, how long has it been open, and who set the expected close date - the buyer or the rep?
4. Does the buyer's side involve formal procurement, security review, or legal redlines?
5. Does your team mandate its own MEDDPICC definitions or scale (second C, Implicate vs. Identify, weighting)? If yes, use theirs and note the mapping in the output.
6. By what date must this scorecard change something - this week's forecast call, the next deal review, or nothing dated?
7. Do you want a one-off read on this deal, or a re-scored record that compounds across reviews?
8. What is your effort ceiling before that date: how many buyer asks you can make, and how much of your champion's internal credit you can spend?
Questions 6-8 re-rank the next actions, not the scoring; which answer moves which action is in Closing the biggest gap.
Evidence collection (only for elements that survive right-sizing), one element per message:
9. For this element: what do you know, and what buyer-side artefact or event backs it - an email, a recorded call, a document you have seen, an intro that was made, a date or number the buyer supplied? "The buyer seems positive" is an answer; it scores accordingly.
## Workflow
1. Run the deal-frame interview questions before any element questions.
2. Right-size (next section). Confirm the element list with the user before scoring anything.
3. Collect, per scored element, the rep's claim and the buyer-side evidence behind it.
4. Score each element on the 0-3 evidence ladder (Scoring section).
5. Challenge every 2 and 3: which artefact or event justifies it? No artefact named → drop the score to 1 and record why. Apply this anti-inflation rule without exception - self-scoring optimism is the documented failure the rule corrects - score levels need objective definitions, or an optimist and a pessimist grade the same deal differently.
6. Apply the two gates, compute the total, band the deal.
7. Name the single biggest gap: a failed gate first; otherwise the gap that buys the most verdict change per unit of effort (Closing the biggest gap).
8. Attach a next action to every element below 3, ordered by that same ratio. Single out the top one, phrased as a buyer-visible test - an ask the buyer either performs or refuses, so the answer is evidence either way.
9. Run the Quality gate; iterate until it passes.
10. Deliver the scorecard using [references/scorecard-template.md](references/scorecard-template.md).
11. If your harness has persistent memory, store the scorecard, its date, and each score's justifying artefact so the next run can re-score and show deltas - deals decay, so re-score at every deal review. Otherwise, end with a short recap the user can paste into their own deal record.
If you can browse the web, you may verify public facts a score depends on (the named Economic Buyer still holds the role; the competitor cited still sells the product); otherwise rely on the user's report and mark those facts "user-reported".
## Right-size before scoring
Full MEDDPICC on every deal is over-engineering; scoring the wrong list is the first inflation. Which letters survive is itself a ranked menu - what scoring the letter costs against the verdict information it returns:
- value on a small deal (verdict change per letter scored): Champion > Implicate the Pain == Metrics > Decision Process > Competition > Decision Criteria > Paper Process
- effort (rep time plus the buyer questions the letter costs, most expensive first): Paper Process > Decision Process > Decision Criteria > Competition > Metrics == Implicate the Pain == Champion
Drop from the low-value, expensive end inward: Paper Process leaves first (the first element dropped in lightweight practice), then Decision Criteria, then Competition.
- Pain == Metrics: they are one answer from two directions: the costed consequence and the number the buyer owns arrive in the same sentence.
- Metrics, Pain and Champion tie at the cheap end and never drop: on a small deal each is a single question, and any two of them are unreadable without the third.
- Economic Buyer sits on neither line: it is a gate, and on small deals it merges into Champion instead of being dropped.
Fit bands:
- Roughly $100K+ contract value, 5+ stakeholders, formal procurement/security/legal path: score all 8.
- ~$25K-$100K, 2-5 stakeholders: drop Paper Process, unless procurement is known to be involved.
- Below ~$25K, 1-2 deciders, no procurement: Metrics + Implicate the Pain + Champion, adding Decision Process if any multi-step approval exists. Here the Economic Buyer and Champion usually collapse into one person - score them as one letter and say so.
- High-ticket B2C: see the B2B and B2C section; right-size aggressively.
Never score an element that was right-sized out; the denominator shrinks with the element list, and a dropped element earns no next action either.
## The eight elements and their evidence bar
| Element | What it means (this skill's chosen definition) | Validated (3) looks like | Assumed (1) looks like |
| ------------------ | ----------------------------------------------------------------------------------- | ---------------------------------------------------------------------- | ------------------------------------------------------- |
| Metrics | The quantified outcome the buyer expects, as a number they own | A figure the buyer supplied in their own words | "They want efficiency"; a rep-built ROI never confirmed |
| Economic Buyer | The one person with discretionary spend authority and veto | A held meeting with the named EB confirming the case | "My contact says the CFO is on board" |
| Decision Criteria | The technical, economic, and relationship standards used to compare options | Criteria the buyer wrote down or confirmed, proof mapped | A feature list or RFP sheet treated as neutral criteria |
| Decision Process | Who decides what, in what order, by when (technical validation + business approval) | A mutual action plan the buyer co-edited | A close date the rep invented; "they said Q3" |
| Paper Process | Everything between verbal yes and signature: legal, procurement, security | A mapped path with named owners and timelines | "Procurement is just a formality" |
| Implicate the Pain | The problem made costed and owned; the cost of inaction | Buyer states the specific costed consequence of not solving | A surface pain with no cost attached |
| Champion | Power + personal stake + sells internally when you're absent | The contact has acted: EB intro, shared criteria, presented internally | A friendly contact who hasn't and maybe can't act |
| Competition | Every alternative: rivals, internal build, other initiatives, inertia | Buyer articulates the alternatives; a win plan positions against each | "We're the only one they're looking at" |
Contested definitions - this skill's positions, to be overridden by the user's team standard when one exists:
- The second C is Competition, including inertia/do-nothing (the methodology publisher's official position); some teams use Compelling Event or Closing instead.
- A compelling event is a thread running through every element, not its own letter; its honest absence is deal intelligence, never something to manufacture.
- Pain means Implicate the Pain: identify the problem, indicate its broader impact, implicate the buyer in the costed consequence.
- The optional "P for Partner" variant is not scored here.
Per-letter evidence exemplars, red flags, and the questions that move a letter up the ladder are in [references/element-evidence-bar.md](references/element-evidence-bar.md) - read it before scoring.
## Scoring
The evidence ladder, per element:
| Score | RAG | Label | Bar |
| ----- | ----- | --------- | --------------------------------------------------------------------------------------------------- |
| 0 | Red | Unknown | No information at all |
| 1 | Red | Assumed | The rep believes it; no buyer-side evidence |
| 2 | Amber | Stated | The buyer said it - in a meeting, call, or email |
| 3 | Green | Validated | The buyer demonstrated it: an intro made, a document seen, a date committed, a number they supplied |
Where the scale comes from: the red/amber/green convention comes from Andy Whyte's book, confirmed as the book's own core scoring mechanism rather than one option among several. The 0-3 evidence ladder is a documented practitioner variant rather than official methodology doctrine, and this skill adopts it and defines each rung precisely above. Per-letter weighting is configurable and practice-dependent, never canonical - this skill weights all elements equally and says so in the output; if the user's team weights differently, use theirs.
Aggregate: total = sum of element scores; percentage = total / (3 × elements scored). Bands:
- **≥80% and both gates passed - Forecast-ready.** Evidence supports committing the deal (anchor: a documented practitioner threshold of 20+/24). Still re-score at every review.
- **55-79% - Real but gapped.** The deal is live but not commit-grade; work the listed gaps before forecasting it.
- **30-54% - Under-qualified.** Not forecastable. Late-stage deals in this band are a coaching trigger (a documented sub-16/24 late-stage trigger).
- **<30% - Not qualified.** Re-qualify from discovery or disqualify. Disqualification is a valid outcome of this skill, not a failure.
Gates - two letters cap the verdict regardless of total. The emphasis on both letters is the methodology's; the cap mechanic is this skill's, not published doctrine:
- Champion ≤1 → verdict capped at Under-qualified. "No Champion, no deal".
- Economic Buyer = 0 → verdict capped at Under-qualified. A champion unwilling or unable to introduce the EB is a major red flag that also questions whether the champion is real.
Place borderline scores with the sourced maturity ladders:
- Metrics: M1 (outcomes delivered for other customers) → M2 (personalized to this prospect) → M3 (validated post-go-live). A buyer-confirmed M2 scores a 3. A recited M1 scores a 1.
- Champion: Contact → Coach → Champion.
- Economic Buyer: unknown → Access (intro committed, not held) → Sponsorship (met, case confirmed). Access scores a 2. Sponsorship scores a 3.
Never present vendor outcome claims ("18% higher win rates / 24% larger deals / 26% shorter cycles", "forecast accuracy to 95%") as expected results - unreplicated vendor claims; the quarantine list is in [references/vendor-claim-quarantine.md](references/vendor-claim-quarantine.md).
## Closing the biggest gap
Every element below 3 generates a next action, and they all compete for one scarce thing: the asks this rep can make of this buyer before the next checkpoint. Rank them by verdict change bought per unit of effort.
Effort here is never money. It is:
- Rep hours.
- Calendar latency.
- How much of the champion's internal credit the ask spends.
- Whether a refused ask can be re-asked.
The eight candidate asks:
- Champion test
- EB meeting
- Pain cost
- Buyer's metric
- Decision-process plan
- Criteria write-down
- Paper-process map
- Do-nothing probe
The ordering below is this skill's, not published doctrine; effort is stated as orders of magnitude rather than figures.
- value (how much closing it moves the band): champion test > EB meeting > pain cost == buyer's metric > decision-process plan > paper-process map > do-nothing probe > criteria write-down
- effort (most expensive first): EB meeting > paper-process map > decision-process plan > criteria write-down > champion test == pain cost == buyer's metric == do-nothing probe
- efficiency (the default order to work): champion test > pain cost == buyer's metric > do-nothing probe > decision-process plan > criteria write-down > EB meeting > paper-process map
Magnitudes, never figures:
- Near-zero: champion test, pain cost, buyer's metric, do-nothing probe - one question on a call already in the diary.
- An hour to draft plus a week waiting for the buyer to return it: decision-process plan, criteria write-down.
- A week and a second introduction: paper-process map.
- A week or more of calendar, plus the champion's credit to arrange it, and it cannot be un-asked: EB meeting.
The champion test leads on value because it is the ask that produces the others: a champion who acts books the EB meeting, and one who refuses is a coach, which caps the deal anyway. The four near-zero asks tie on effort because none needs a new meeting, a new document, or a second introduction.
What this order starves: the EB meeting and the paper-process map. The two highest-value asks in the deal sit at the bottom of the efficiency line every round, because the element that most changes the verdict is usually the most expensive to validate. Promote them anyway when:
- Economic Buyer scores 0. The gate caps the verdict, so no other validation can lift the band - the EB meeting goes first regardless of ratio.
- The deal is late-stage, procurement is known, or a verbal yes is close. The paper-process map goes first; after the yes it is no longer a cheap fix, it is a stall.
Delete, never demote:
- An element right-sized out has no next action at all.
- An effort ceiling that forbids spending the champion's credit removes the EB meeting and the procurement introduction from the menu, rather than parking them at the bottom where they silently return as scope.
- A team standard that replaces the second C with Compelling Event deletes the do-nothing probe with it.
Say in the scorecard which asks you deleted and why.
Re-rank on what you already know, and record which answer moved which ask:
- A checkpoint inside the week (interview Q6) promotes the near-zero asks and demotes the EB meeting and the paper-process map - neither lands in time.
- A compounding record (Q7) promotes the decision-process plan and the paper-process map; both leave an artefact that keeps scoring at every re-score, which a one-off answer does not.
- A ceiling of one buyer ask (Q8) collapses the menu to the top of the efficiency line.
- A champion who has already acted once makes the EB meeting far cheaper than this order assumes - promote it.
- A team-mandated methodology or weighting (Q5) overrides the ordering outright; use theirs.
This order is a default, not a law. It shifts with the deal and with who executes it: a rep with a standing relationship to the economic buyer pays near-zero for a meeting that costs a first-year rep a week and a favour.
## Invocation and expected output
Typical invocations:
- "Score this deal against MEDDPICC: 200-seat rollout at a logistics company, $180K ACV, proposal stage - here's what we know."
- "MEDDPICC check before Thursday's deal review. Ask me your questions one at a time."
- "Is this deal real? Single-founder buyer, $9K contract, says he'll sign this month."
Deliver one scorecard (full fill-in template in [references/scorecard-template.md](references/scorecard-template.md)):
```
MEDDPICC SCORECARD - <deal>, <date>, elements scored: <list>
Per element : score 0-3 + RAG, the claim, the evidence artefact
(or "none - score dropped"), the next action and what it
costs, best ratio first
Gates : Champion / Economic Buyer status; cap applied or not
Total : X/<max> (NN%), band, one-line verdict in forecast terms
Biggest gap : one element, why it beat the others on verdict change per
unit of effort, and the buyer-visible test that closes it
Deferred : the high-value asks the order starved, what would promote
them, and any ask deleted by a constraint
Challenged : every score knocked down, and why
Definitions : second-C / Pain choices used; weighting (equal by default)
Re-score : the event or date that triggers the next scoring pass
```
## B2B and B2C
- MEDDPICC is B2B-native: its own publisher positions it for mid-market and enterprise B2B, and consumer selling is outside the methodology's stated scope. Everything below about B2C is a structural analogy rather than established practice; say so in every B2C scorecard you produce.
- Identical in both, explicitly:
- The evidence axis.
- The 0-3 ladder.
- The anti-inflation rule.
- The gates.
- The workflow.
- The quality gate.
- Only the cast changes.
- High-ticket considered B2C mapping, all of it analogy:
- Economic Buyer → whoever controls the money and holds veto: spouse or partner, a financing parent, the lender whose approval gates the purchase.
- Champion → the enthusiastic household member selling the skeptical one when the salesperson is absent.
- Decision Process → who must agree, in what order: viewing, inspection, second opinion.
- Paper Process → financing approval, inspection, appraisal, contingencies, closing.
- Competition → rival options plus the powerful "do nothing / wait".
- Metrics → quantified household stakes: monthly payment, resale value, energy savings.
- Pain → the costed and emotional price of the status quo.
- Right-size B2C aggressively: usually Metrics + Pain + Champion + Paper Process as financing; a full eight-letter scorecard on a household purchase is process theatre. Paper Process is the one letter that inverts the drop order here - an unstarted loan application, not an unmapped procurement path, is what actually stops a household purchase, so the starved ask is promoted by default in B2C.
- A worked B2C adaptation is in [references/worked-examples.md](references/worked-examples.md).
## Quality gate
Score the drafted scorecard against all nine checks. Pass threshold: 9/9. Iterate - fix and re-check - until nothing fails.
1. Every scored element carries a claim, a score, and either a named artefact/event or an explicit "none".
2. No 2 or 3 stands without its artefact; every knocked-down score appears under Challenged with the reason.
3. The element list matches the right-sizing decision, and the denominator matches the element list.
4. Both gates were checked; any cap applied is stated in the verdict line.
5. The biggest gap is exactly one element; the scorecard says why it beat the others on verdict change per unit of effort, or which promotion condition overrode the ratio; and its next action is a buyer-visible test, not rep homework ("update the deal record" fails; "ask the champion to book the EB meeting this week" passes).
6. Every element below 3 has a next action, ordered by that same ratio, with the starved asks named under Deferred; no action is attached to a 3.
7. The band's meaning is stated in forecast terms, not as a grade.
8. The chosen definitions (second C, Pain) and the weighting are stated, with a note to align with the team's own standard if one exists.
9. Every rule the scorecard leans on is marked as published methodology or as this skill's own, and no vendor claim appears as an expected outcome.
These nine checks are deliberately unranked: the threshold is 9/9, so there is no first one to run and no last one to skip, and an order here would only invite trading one away.
## KPIs and measurement
- Per deal, the scorecard worked if:
- At least one score was knocked down for missing evidence.
- The top next action is a buyer-visible test.
- Re-scores move - a score that never changes usually means nobody talked to the buyer that week.
- Across deals, the measures are ranked too - value is how much a measure can prove the rubric wrong, effort is how long you wait before it says anything:
- value: band vs. realized outcome > commit-to-close forecast accuracy > no-decision loss rate > slipped-deal count
- effort: band vs. realized outcome > commit-to-close forecast accuracy > slipped-deal count == no-decision loss rate
Slippage and no-decision rate tie at near-zero because both are already columns in any pipeline report. Read them weekly, but they only ever hint.
Band vs. realized outcome is the starved measure - a standing job that pays out only after a full quarter of closed deals - and the sole one that can falsify the rubric, so instrument it from day one anyway. Late-stage slippage bleeds win rate fast (as relative lifts).
- Adoption check: if scorecards get filled but forecast accuracy does not move, the evidence definitions are too loose. That is MEDDPICC theatre - fields completed to satisfy a manager rather than to qualify - the exact documented failure this skill exists to prevent.
## Common failure modes
| Failure | Fix |
| ---------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------ |
| Happy ears - buyer politeness read as validation | Demand the artefact; politeness is not an event |
| Self-scoring inflation - optimism graded as evidence | The anti-inflation rule: no named artefact, no 2 or 3 |
| Presence, not evidence - "we have a champion" scored green | Score what the person has done, not that they exist |
| Coach mistaken for champion | Test with an ask - EB intro, internal criteria, presenting to the committee; hesitation means coach |
| Feature list or RFP mistaken for decision criteria | Criteria count only when buyer-confirmed and covering technical, economic, and relationship dimensions |
| Paper process discovered after the verbal yes | Score it early on large deals; "procurement is just a formality" is a 1 |
| Inertia ignored as competition | "Do nothing" is the most common winner; no do-nothing analysis caps Competition at 2 |
| Score used as a report card, not an action generator | Every gap gets a next action, ordered by verdict change per unit of effort; the real output is the ranked to-do list, not the number |
| Biggest gap picked on value alone - the EB meeting chosen the day before a forecast call | Rank by ratio, not importance; promote a starved ask only on its stated condition |
| Single-threaded deal scoring green | One contact supplying the evidence for every letter is itself fragility - flag it on the scorecard |
| Over-qualifying small deals | Right-size first; all eight letters on a $9K single-decider deal is process theatre |
## Reference
- [references/element-evidence-bar.md](references/element-evidence-bar.md) - per-letter definitions, evidence exemplars, assumption red flags, ladder-climbing questions.
- [references/scorecard-template.md](references/scorecard-template.md) - the fill-in-ready scorecard.
- [references/worked-examples.md](references/worked-examples.md) - a full B2B enterprise scorecard, an annotated early-stage negative example, and a high-ticket B2C adaptation.
- [references/vendor-claim-quarantine.md](references/vendor-claim-quarantine.md) - the circulating MEDDPICC figures that are vendor marketing, and how to handle them.
- mbfinotti/sales-skills@sales-discovery-questions - generates the interview questions; this skill only grades what came back.
- mbfinotti/sales-skills@deal-value-calc - builds the ROI/business case behind Metrics; here the buyer's number is only checked as evidence.
- mbfinotti/sales-skills@sales-hiring - ramp acceleration via shared qualification language.