mbfinotti/sales-skillsCheck passed
SKILL DETAIL
deal-champion-mapping
mbfinotti/sales-skills/deal-champion-mapping
Builds an evidence-graded stakeholder map for one active deal from titles, call notes, email and meeting activity, and CRM contact roles, naming the likely champion, economic buyer, and blockers, each with a confidence level tied to observed behaviour rather than title. Covers B2B buying committees and high-consideration B2C households. Use whenever the user mentions a champion, economic buyer, power map, deal org chart, multi-threading, single-threaded risk, or "who actually decides", even without the word stakeholder. Do NOT use for qualification scoring (mbfinotti/sales-skills@meddpicc-scorecard) or scanning notes for general risk (mbfinotti/sales-skills@deal-red-flags).
Installs · 187View source
Installation
npx skills add https://github.com/mbfinotti/sales-skills --skill deal-champion-mapping
Skill files
SKILL.md
Last synced · Sep 15, 2026
evals/evals.json›
{
"skill_name": "deal-champion-mapping",
"evals": [
{
"id": 1,
"prompt": "Need a read on the Caldwell Nutrition account before my manager's pipeline review Thursday. It's a $140K ACV deal, they're about 2,400 employees, we're in evaluation. My main contact is Renata Vogel, Sr. Manager of Supply Planning - she replies within about ten minutes every time, has joined all five of our calls, and told me on call 3 'I'm the decision maker on this, don't worry about anyone else.' Also on the thread: Tomas Iriarte, VP Finance (cc'd twice, never spoken), and Aiko Brennan, a supply planning analyst who's been on two calls and loves the product. I haven't asked Renata for anything yet. Our platform stores their supplier contact records. Who's my champion and who actually signs off?",
"expected_output": "A stakeholder map that grades Renata's responsiveness as cheap behaviour rather than champion proof, caps the VP Finance economic-buyer guess at title-level confidence, flags the economic buyer as unverified with a surfacing action, sets the next ask at the bottom rung of the ladder, and hypothesises a security reviewer nobody has met.",
"files": [],
"expectations": [
"Renata's fast replies and meeting attendance are graded as Tier 2 low-cost behaviour, not as champion evidence.",
"Renata's 'I'm the decision maker' statement is graded as a Tier 3 stated claim and is not treated as proof.",
"Renata is not assigned Champion at High confidence; she is held at Medium or lower, or labelled untested, coach, or champion-candidate.",
"The economic-buyer hypothesis for Tomas Iriarte rests on title only, is labelled Tier 4, and is capped at Low confidence.",
"The output states that at roughly 2,400 employees a VP-level title often sits two levels away from the budget, weakening title inference.",
"The economic buyer is either explicitly flagged unknown or named only as a Low-confidence hypothesis, with a named surfacing action attached.",
"At least one role-identifying question is proposed, such as who approved the last similar purchase, who holds the budget and could redirect it, or who could say no when everyone else says yes.",
"The next ask on the champion ladder is rung 1 - share internal evaluation criteria, competing vendors, or budget - justified by Renata being untested.",
"Because the product stores supplier contact records, a security or data-handling reviewer is hypothesised even though no such person has appeared.",
"At least one blocker hypothesis names a function that has not been on any call.",
"Aiko Brennan is classified as a user or cheerleader rather than as an influencer with organisational pull.",
"The engaged-contact count is compared against a segment benchmark and a refresh trigger is stated."
]
},
{
"id": 2,
"prompt": "We're bidding on a formal RFP from the Alderbrook Regional Transit Authority - public agency, five-year contract, they published the requirements doc and everything routes through their procurement office. Evaluation stage, responses due in six weeks. My contact is Dev Okonkwo, Deputy Director of Operations; he ran the discovery workshop and has answered every email within a day. Their CFO, Marisol Paz, signs anything over $250K. Who's my champion and what do I ask Dev for next to lock this up?",
"expected_output": "A map that removes the criteria-and-competitors ask entirely because of the public RFP, sets the next ask at a surviving rung, separates the signatory from the economic buyer, and works the blocker set in efficiency order.",
"files": [],
"expectations": [
"Asking Dev to share internal evaluation criteria, competing bidders, or the budget is deleted from the plan, not merely deprioritised.",
"The deletion reason names the formal RFP or public procurement context and the risk that requesting competing-bid information can disqualify the bid.",
"The recommended next ask is a rung that survives the deletion - a forwardable rep-written summary, or an introduction to the economic buyer - and it is named as a rung.",
"The co-built business case is not deleted for stage reasons, because the deal is at evaluation rather than discovery.",
"The output does not recommend asking Dev who else is bidding or what the internal scoring criteria are.",
"Dev's responsiveness and workshop facilitation are graded no higher than Tier 2 in the absence of a capital-costing ask.",
"Marisol Paz is classified as signatory or budget holder rather than as the confirmed economic buyer.",
"The 'power to say yes when others say no, and no when others say yes' test is named as what decides the economic buyer.",
"Procurement appears as a mapped function, with the note that engaging it early is near-zero effort only if done before the paper stage.",
"Legal appears in the blocker set with a pre-emptive standard-terms play.",
"At least one blocker hypothesis covers a function not yet on any call.",
"A refresh trigger is stated."
]
},
{
"id": 3,
"prompt": "Stuck on the Peregrine Dental Group deal, $95K, proposal stage, closes end of quarter. Everything runs through Hollis Marchetti, their Director of Clinical Ops. She's great - takes every call, gave me their rollout timeline. Twice now I've asked her to set up 20 minutes with their CFO, Warren Amadi, and both times she said 'let me handle him, he trusts my recommendation.' She is genuinely the only person at that account I have a relationship with. I found Warren on LinkedIn and I'm thinking of just messaging him directly with a short note. Thoughts, and can you map this out?",
"expected_output": "A map that downgrades Hollis on two deferrals, refuses the LinkedIn bypass and deletes it because the deal is single-threaded, runs the deferral readings in their ranked order starting with the value-exchange one-pager, and flags the coverage gap.",
"files": [],
"expectations": [
"Hollis is downgraded from champion to coach, or explicitly withheld from champion status, on the basis of two deferrals of a capital-costing ask.",
"Messaging Warren directly around Hollis is rejected, not merely cautioned against.",
"The rejection states that going around the champion can convert her into a detractor.",
"The bypass is deleted specifically because the deal is single-threaded, rather than kept as a fallback if the countermoves fail.",
"The readings of a deferred economic-buyer introduction are applied or listed, each paired with its countermove.",
"The first countermove run is the value exchange: a short economic-buyer-ready business case Hollis can present herself and look good doing it.",
"The ordering logic behind the countermoves is stated explicitly - access bought per rep hour and per unit of the contact's capital - rather than left implicit in the order they appear.",
"A give-to-get is offered as a softener: something Hollis wants, traded for the introduction, stated as a condition up front.",
"Single-threading is flagged as a coverage gap, with a named function to expand into.",
"The deal being over $50K and single-threaded is called structurally behind.",
"Warren is not recorded as a confirmed economic buyer; his assignment carries its evidence tier and confidence.",
"Champion risk is answered on all three points - redundancy candidate, executive sponsor, job-change detection - with any absent one marked as a gap."
]
},
{
"id": 4,
"prompt": "Mapping blockers on Tallgrass Freight Systems - $110K, mid-market, evaluation. Our product optimises trailer routing off their shipment manifests; no personal data, no consumer records, and they're not in a regulated industry. On the last call their CTO, Idris Vantol, said 'we'll need to run this through IT' and asked three questions about how we'd integrate with their TMS. My champion Simone Auer offered to set up a two-week proof of concept with their integration team, which sounds like a fast way to prove value. Their general counsel apparently reviews every vendor MSA. Give me the blocker map and tell me what order to work them in.",
"expected_output": "A blocker map ordered by veto risk removed per unit of work, with the compliance row deleted for this buyer, the proof of concept held unless a promoting condition fires, and effort given in orders of magnitude.",
"files": [],
"expectations": [
"The blocker plays are ordered by efficiency - veto risk removed per unit of work - and that ordering is stated out loud rather than implied by row order.",
"Legal ranks first, with the pre-emptive play of sending standard terms before they are requested, at near-zero effort.",
"The compliance and privacy row is deleted for this deal, with the reason that the buyer is unregulated and the product touches no personal data.",
"The deleted compliance row is removed from the plan rather than carried at the bottom of the table.",
"The proof of concept is held rather than adopted as the next step, and the output says the efficiency order is what starves it.",
"A promoting condition for the proof of concept is stated: either the technical reviewer holds the real veto and has raised an integration objection the documents cannot answer, or the differentiation is only visible inside the buyer's own environment.",
"The output does not adopt the two-week proof of concept as the default next step merely because the champion proposed it.",
"Idris Vantol's integration questions are treated as a blocker signal rather than as engagement or support.",
"The output distinguishes the senior technical executive from whoever runs the actual technical review, noting the reviewer can hold informal veto.",
"Effort is expressed in orders of magnitude - near-zero, an hour, a week, a quarter - and never as a currency amount.",
"At least one blocker hypothesis names a function that has not appeared on any call.",
"Blockers are not ordered by the seniority of the person who would object."
]
},
{
"id": 5,
"prompt": "We sell residential solar. Booked a consult at a house in Fairhaven - Jodi Streit called us in, she's read everything, took the full 90-minute site walk and asked about panel degradation curves. Her husband Petar sat in for twenty minutes and said 'it's a lot of money for a roof that's eight years old.' Jodi mentioned her father helps them with big financial decisions and 'will have opinions.' System is about $34K before incentives and they'd finance it. Their 19-year-old still lives at home. I'm sending the proposal Friday - how do I read this household?",
"expected_output": "A household map using decision-making-unit roles, with the B2B-only blocker functions deleted, the father mapped as an absent hypothesis, a capital-costing ask for Jodi before Friday, and the same evidence tiers and confidence rules applied unchanged.",
"files": [],
"expectations": [
"The household is mapped with decision-making-unit vocabulary: initiator, influencer, decider, buyer, user, gatekeeper.",
"Jodi is identified as the champion or initiator, and the output names what would prove it.",
"Petar is treated as a potential detractor or blocker rather than as a neutral participant.",
"The decider role is mapped onto the economic-buyer slot, and the output says whether the decider is confirmed or unknown.",
"Procurement, legal, security and compliance rows are deleted for this household rather than given invented household equivalents.",
"The job-change-detection row is dropped, and champion risk is restated as what happens if the enthusiastic partner cools before the decider engages.",
"The economic, political and incumbent-loyalist blockers are kept, each with a household reading.",
"Doing nothing, or keeping the current arrangement, is named as a live competitor.",
"A capital-costing ask is proposed that costs Jodi real effort, with a deadline before Friday's proposal.",
"The evidence tiers are applied unchanged: the 90-minute site walk and the degradation question are graded, and confidence follows the same rules as a B2B map.",
"The output states explicitly that this is the high-consideration household adaptation rather than leaving the transfer implicit.",
"The father is mapped despite never having been present, as a hypothesis rather than as an observation."
]
},
{
"id": 6,
"prompt": "Quick sanity check on Brightmoor Analytics - $180K deal, mid-market, we're at proposal. I've got nine named contacts in the CRM and I've emailed all nine at least once, so we're well past the 4-7 benchmark my manager quotes. Realistically only Kenji Haruna, their Head of Revenue Ops, actually replies; two others replied once in March and nothing since. My manager wants me at fifteen contacts before the forecast call. Am I covered, and how many more should I add?",
"expected_output": "A coverage assessment that counts real relationships rather than contacted names, calls the deal single-threaded, refuses the push to fifteen, cites the falling-consensus counter-current, and names specific unmapped roles worth adding.",
"files": [],
"expectations": [
"Coverage is counted in real relationships rather than named contacts or email addresses, and the real-relationship count here is one.",
"The deal is flagged single-threaded despite nine contacted names.",
"The benchmark applied is the $50K-$250K band of roughly ten engaged contacts, not the 4-7 mid-market figure the user quoted.",
"Benchmarks are described as directional and correlational rather than as a quota or a target.",
"Pushing to fifteen contacts in order to hit a number is rejected.",
"The output cites the counter-current that purchase likelihood falls as the buying group grows - roughly 81% with one decision maker, 55% with two, the mid-30s at six.",
"Any multithreading win-rate lift quoted is labelled vendor-data and correlational, never causal.",
"Contacts are recommended only to cover named unmapped roles, and at least two such roles are named.",
"The deal being over $50K and single-threaded is called structurally behind.",
"Kenji's evidence carries a tier and his role assignment carries a confidence level.",
"Champion risk is answered, including whether a redundancy candidate exists.",
"A refresh trigger is stated."
]
},
{
"id": 7,
"prompt": "Building the stakeholder picture for Vantour Systems, $70K, evaluation stage. Two things I keep seeing quoted that I want to build around: that 40% of so-called champions fail the champion test, and that Gartner found B2B buyers consume 17 pieces of content before they decide. My contact Adaeze Nwankwo, Head of Customer Support, forwarded my one-pager to her director last week, same day I sent it. Nobody above her has spoken to us. Map it out and tell me what those numbers mean for how hard I push.",
"expected_output": "A map that refuses to build on the folklore figure, corrects the misattributed Gartner statistic and supplies the real committee figure, grades the same-day forward as a passed rung, and climbs exactly one rung from there.",
"files": [],
"expectations": [
"The 40% champion-failure figure is labelled practitioner folklore with no traceable methodology and is not used as the basis for any recommendation.",
"The '17 pieces of content' claim is identified as a misattribution of Gartner's research.",
"The correct reading is supplied: Gartner's 17 is the share of buying time buyers spend meeting all potential suppliers combined, roughly 17%.",
"Gartner's actual committee figure is supplied instead - a median buying group of six to ten decision makers.",
"Adaeze forwarding the one-pager the same day is graded as a passed rung with follow-through inside its stated timeframe.",
"The forward is graded as Tier 1, and the output notes it proves willingness rather than standing.",
"The next ask is an introduction to the economic buyer or an executive sponsor, named as a rung.",
"The already-passed forwardable-summary rung is not repeated, and the output states that a second cheap ask buys no new evidence.",
"The largest ask - co-building and presenting the business case - is not selected as the next step.",
"The economic buyer is flagged unknown with a named surfacing action, since nobody above Adaeze has engaged.",
"Every vendor-sourced statistic appearing on the map is labelled correlational.",
"At least one blocker hypothesis names a function not yet on any call."
]
},
{
"id": 8,
"prompt": "Early days on Hartsfield Bioworks - we're in discovery, maybe $60K, no proposal yet and no committee formed. Three weeks ago Priyanka Deol, their Lab Operations Manager, sent me their internal scoring sheet for the two vendors they're looking at, which I'm fairly sure she wasn't supposed to share. Separately, their CFO Gideon Marsh and I worked together at my last company; I can text him any time. What should I ask Priyanka for next?",
"expected_output": "A next-ask recommendation that grades the leaked scoring sheet as the bottom rung already passed, deletes the introduction rung because the CFO relationship already exists, deletes the business-case rung because the stage has no committee, and lands on the forwardable summary.",
"files": [],
"expectations": [
"Priyanka sharing the internal scoring sheet is graded as Tier 1 evidence and identified as the criteria-share rung already passed with follow-through.",
"That rung is not repeated, and the output states that a passed rung buys no new evidence.",
"The economic-buyer introduction rung is deleted because the existing CFO relationship means the introduction is already paid for.",
"The co-built business case rung is deleted because the deal sits at discovery with no committee meeting to present at.",
"The recommended next ask is the forwardable rep-written summary or a customer reference call, named as a rung.",
"Deleted rungs are removed from the plan rather than parked at the bottom of the ladder.",
"The output says which known fact moved which rung.",
"A forward-commitment phrasing is proposed that secures the commitment live on a call before the summary is written or sent.",
"Gideon Marsh is not automatically recorded as the confirmed economic buyer, and the 'yes when others say no' test is applied to him.",
"The existing CFO relationship is used to re-rank the ladder rather than as a reason to bypass Priyanka.",
"A blocker hypothesis is produced for at least one function not yet on any call.",
"Priyanka's role assignment carries an evidence tier and a confidence level."
]
},
{
"id": 9,
"prompt": "Novaris Health, $220K, evaluation, and their board meets in nine days - if we're not on that agenda it slips two quarters. My contact Bethan Ortiz, Director of Care Coordination, already spent goodwill once this month getting us a security exception fast-tracked, and she told me last Tuesday 'I can't keep asking people for favours on this.' Their architecture team floated a sandbox pilot. What do I ask for and in what order?",
"expected_output": "A re-ranked ask plan where the nine-day deadline deletes the long-latency business case, the exhausted political capital deletes the sandbox pilot, and the recommendation is a near-zero same-week ask with the constraint behind each change named.",
"files": [],
"expectations": [
"The nine-day board deadline is used to re-rank the ask ladder, and the output says so explicitly.",
"The co-built business case is deleted because its exec-review latency exceeds the nine days available.",
"The economic-buyer introduction is flagged as carrying roughly a week of calendar latency, and is either deleted or kept only with that latency stated against the deadline.",
"The sandbox pilot or proof of concept is deleted on the effort ceiling, because it spends both the buyer's engineering time and Bethan's capital to schedule.",
"The recommended ask is a near-zero, same-week one, named as a rung.",
"Bethan's fast-tracked security exception is graded as Tier 1 behaviour under cost.",
"The output states that a contact who has already spent capital once does not start back at the bottom rung.",
"The output names which constraint moved which rung - the deadline for one, the effort ceiling for another.",
"Deleted asks are removed from the plan rather than demoted to the bottom of a list.",
"The output does not recommend spending Bethan's remaining capital on more than one ask before the board date.",
"A refresh trigger tied to the board meeting is stated."
]
},
{
"id": 10,
"prompt": "Deal: Kirkholm Manufacturing, $130K, paper stage. Their VP Engineering, Anselm Roux, signs all our contracts - he's signed two other vendor agreements this year - so he's obviously the economic buyer. The budget line for tooling software sits with Delphine Kovac, Director of Platform. I also want a MEDDPICC score out of this and a full discovery question list for my next call with Delphine. Map the stakeholders.",
"expected_output": "A map that separates signatory and budget holder from the economic buyer, leaves the economic buyer unproven with a surfacing action, supplies only the role-identifying questions, and declines both the MEDDPICC score and the full discovery set by pointing at the right skills.",
"files": [],
"expectations": [
"Anselm is classified as signatory, and the output states that a signatory may be a delegate executing the economic buyer's decision.",
"Anselm is not recorded as the confirmed economic buyer on the strength of having signed other vendor agreements.",
"Delphine is classified as budget holder, with the note that owning a budget line does not imply authority to redirect it to this purchase.",
"The 'power to say yes when others say no, and no when others say yes' test is applied to decide who the economic buyer is.",
"The economic buyer is flagged unknown, or held at no more than Low or Medium confidence, with a surfacing action attached.",
"At most three role-identifying questions are supplied, rather than a full discovery question set.",
"The request for a full discovery question list is redirected to the discovery-questions skill rather than fulfilled here.",
"No MEDDPICC score, letter grade, band, or acronym-based qualification verdict is produced.",
"The MEDDPICC request is redirected to the MEDDPICC scorecard skill by name.",
"At the paper stage, procurement and legal both appear on the blocker map with their plays.",
"Every named stakeholder carries a role, an evidence tier with the evidence itself, a confidence level, and a next action."
]
}
],
"trigger_queries": [
{ "query": "who's the real decision maker on the Ellsworth deal?", "should_trigger": true },
{ "query": "map the stakeholders on this opportunity", "should_trigger": true },
{ "query": "is Renata a champion or just friendly?", "should_trigger": true },
{ "query": "build me a power map for the Corveline account", "should_trigger": true },
{ "query": "here's my contact list and three call transcripts - who blocks this?", "should_trigger": true },
{ "query": "I think this deal is single-threaded, how bad is that", "should_trigger": true },
{ "query": "my only contact keeps refusing to introduce me to the CFO", "should_trigger": true },
{ "query": "who signs off on this and who can kill it", "should_trigger": true },
{ "query": "deal org chart please", "should_trigger": true },
{ "query": "we've only ever spoken to one person at this account", "should_trigger": true },
{ "query": "how do I tell if she'll actually go to bat for us internally", "should_trigger": true },
{ "query": "my champion just resigned, what now", "should_trigger": true },
{ "query": "buying committee map for a $200K enterprise deal", "should_trigger": true },
{ "query": "who else should I be talking to at this account", "should_trigger": true },
{ "query": "I don't know who controls the budget here", "should_trigger": true },
{ "query": "grade my contacts on how much evidence I actually have on each of them", "should_trigger": true },
{ "query": "multi-threading plan for the Wexley renewal", "should_trigger": true },
{ "query": "we lost the last one to someone in security we never met - how do I avoid that again", "should_trigger": true },
{ "query": "what should I ask my main contact for next to prove she's a champion", "should_trigger": true },
{ "query": "is the guy who signs the contract the same as the economic buyer?", "should_trigger": true },
{ "query": "who's the silent detractor on this deal", "should_trigger": true },
{ "query": "map the household before I send the solar proposal", "should_trigger": true },
{ "query": "the husband loves it, the wife hasn't said a word - how do I read this", "should_trigger": true },
{ "query": "my contact says she's the decision maker, should I believe her", "should_trigger": true },
{ "query": "I need to figure out the politics inside this account", "should_trigger": true },
{ "query": "who's going to say no at the last minute", "should_trigger": true },
{ "query": "we have eight contacts but only one ever replies - are we covered", "should_trigger": true },
{ "query": "champion test for the Aldermoor deal", "should_trigger": true },
{ "query": "should I go around my contact to reach the CFO?", "should_trigger": true },
{ "query": "what's my coverage look like on this opportunity", "should_trigger": true },
{ "query": "list who I'm missing on this deal", "should_trigger": true },
{ "query": "how many people should realistically be involved in a deal this size", "should_trigger": true },
{ "query": "identify the economic buyer from these call notes", "should_trigger": true },
{ "query": "is my main contact a coach or a champion", "should_trigger": true },
{ "query": "how do I get to the person who actually controls the budget", "should_trigger": true },
{ "query": "risk assessment on relying on one person at this account", "should_trigger": true },
{ "query": "who has veto power here", "should_trigger": true },
{ "query": "do I have an exec sponsor on this deal or not", "should_trigger": true },
{ "query": "stakeholder map for the Brixton Foods opportunity, evaluation stage", "should_trigger": true },
{ "query": "he takes every meeting but nothing ever moves - what's going on", "should_trigger": true },
{ "query": "i pasted my crm contact roles below, tell me what's real and what's a guess", "should_trigger": true },
{ "query": "who at this account could quietly kill it without ever joining a call", "should_trigger": true },
{ "query": "before I send the proposal, who haven't I met", "should_trigger": true },
{ "query": "what's the next favour I can reasonably ask my contact for", "should_trigger": true },
{ "query": "should I ask for the intro to the CFO now or wait", "should_trigger": true },
{ "query": "we're about to go to paper and I've never met anyone above director level", "should_trigger": true },
{ "query": "kitchen remodel lead - the wife is sold, her mother is paying. how do I play it", "should_trigger": true },
{ "query": "rank the people on this deal by how much they can hurt me", "should_trigger": true },
{ "query": "my contact shared their internal vendor scorecard with me - what does that tell you", "should_trigger": true },
{ "query": "how do I know my champion is real and not just nice", "should_trigger": true },
{ "query": "who do I need on my side to win Hollingbrook", "should_trigger": true },
{ "query": "everyone I've met likes us and the deal still isn't moving", "should_trigger": true },
{ "query": "tell me which of these five contacts actually matters", "should_trigger": true },
{ "query": "score this deal against MEDDPICC", "should_trigger": false },
{ "query": "run a MEDDIC qualification on the Ferrow account", "should_trigger": false },
{ "query": "check my deal notes for red flags", "should_trigger": false },
{ "query": "is this deal at risk? scan the notes and grade what you find", "should_trigger": false },
{ "query": "write the discovery questions for tomorrow's call", "should_trigger": false },
{ "query": "build a question set for a 30-minute qualification call", "should_trigger": false },
{ "query": "build the ROI business case for this deal", "should_trigger": false },
{ "query": "what payback period should we show them", "should_trigger": false },
{ "query": "plan my concessions before the pricing negotiation", "should_trigger": false },
{ "query": "what can I trade instead of discounting", "should_trigger": false },
{ "query": "score this call transcript against our rubric", "should_trigger": false },
{ "query": "review my rep's call recording and tell me what to coach", "should_trigger": false },
{ "query": "write the recap email from today's meeting", "should_trigger": false },
{ "query": "turn these notes into a follow-up email with agreed next steps", "should_trigger": false },
{ "query": "how many touches should my outbound sequence have", "should_trigger": false },
{ "query": "design the cadence for inbound lead follow-up", "should_trigger": false },
{ "query": "write me a cold call opener for CFOs", "should_trigger": false },
{ "query": "what do I say in the first ten seconds when they pick up", "should_trigger": false },
{ "query": "test these five subject lines and tell me which to send", "should_trigger": false },
{ "query": "my emails are landing in spam, what's wrong with my domain setup", "should_trigger": false },
{ "query": "handle the 'we're happy with our current vendor' objection", "should_trigger": false },
{ "query": "they said we're too expensive, what do I say", "should_trigger": false },
{ "query": "define our ICP from last year's closed-won deals", "should_trigger": false },
{ "query": "which firmographic signals should our account fit score use", "should_trigger": false },
{ "query": "set the tier cutoffs and coverage model for our accounts", "should_trigger": false },
{ "query": "how many accounts per rep in tier 1", "should_trigger": false },
{ "query": "design the comp plan for our new AE team", "should_trigger": false },
{ "query": "what pay mix should an SDR role have", "should_trigger": false },
{ "query": "how much pipeline do we need to hit next quarter's quota", "should_trigger": false },
{ "query": "what coverage ratio should we be running", "should_trigger": false },
{ "query": "set quotas for the new fiscal year", "should_trigger": false },
{ "query": "TAM SAM SOM for the European mid-market", "should_trigger": false },
{ "query": "should we restructure into pods or stay on the island model", "should_trigger": false },
{ "query": "what SDR to AE ratio for a team our size", "should_trigger": false },
{ "query": "are we product-led or sales-led, and should we change", "should_trigger": false },
{ "query": "which sales podcasts and newsletters should I be following", "should_trigger": false },
{ "query": "how do I get promoted from SDR to AE", "should_trigger": false },
{ "query": "design the interview loop for hiring AEs", "should_trigger": false },
{ "query": "find three personalization angles for this prospect", "should_trigger": false },
{ "query": "they raised a Series B last week, what's my hook", "should_trigger": false },
{ "query": "write the weekly stakeholder update for our project sponsors", "should_trigger": false },
{ "query": "build a RACI matrix for this internal migration project", "should_trigger": false },
{ "query": "identify internal champions for our change-management rollout", "should_trigger": false },
{ "query": "build a developer champions program for our API", "should_trigger": false },
{ "query": "design a partner ecosystem map for our integrations", "should_trigger": false },
{ "query": "our top partner's exec sponsor went quiet on the co-sell agreement, how do I re-engage", "should_trigger": false },
{ "query": "clean up the contact role fields in our CRM", "should_trigger": false },
{ "query": "build a lead scoring model", "should_trigger": false },
{ "query": "customer health score to predict churn", "should_trigger": false },
{ "query": "which customers should get an executive sponsor in our CS program", "should_trigger": false },
{ "query": "who are the decision makers we should be targeting in the mid-market segment", "should_trigger": false },
{ "query": "find the best deal on a hotel in Lisbon for the customer conference", "should_trigger": false },
{ "query": "write a job description for a head of sales", "should_trigger": false }
]
}
references/evidence-and-benchmarks.md›
# Evidence base and benchmarks
Everything a map cites lives here with its source and caveats. Labels: **verified** (confirmed at a named source), **second-hand** (only via another publisher), **vendor-data** (a vendor's own correlational platform dataset, never independently replicated), **inference** (reasoned from the labelled figures above, not itself reported). No vendor figure below is causal; engagement co-occurs with winning, and the arrow is unproven.
## Buying-committee size
- Gartner: "the median B2B buying group involves six to 10 decision makers, each armed with four or five pieces of information" - **verified** to Gartner's _The New B2B Buying Journey_ / 2017 Digital B2B Buyer Survey (n=750). Median, complex solutions - not all B2B.
- Misquotation warning: the circulating "buyers consume 17 pieces of content" is a **verified misattribution** - Gartner's real "17" is that buyers spend only 17% of their buying time meeting with all potential suppliers combined.
- CEB: 5.4 stakeholders (2014 survey of 3,000+ B2B stakeholders, published in _The Challenger Customer_), rising to 6.8 by 2016 - **verified** as CEB's figures. The "6.8, HBR 2017" citation is the same data republished.
- Forrester _The State of Business Buying_ (2026): the typical buying decision includes 13 internal stakeholders and 9 external influencers - **verified** as Forrester's claim.
- 6sense: average buying group ~10-11 people, rising with selling price - **vendor-data** (2023 survey of 1,000+ practitioners/buyers).
- These figures differ in scope (median vs mean; "decision makers" vs "anyone involved"), not in direction: committees have grown and scale with deal value.
## Multithreading and EB engagement
- Gong (1.8M new-business deals, 2024 analysis): "multi-threading boosts win rates by an average of 130% in deals over $50K"; large strategic deals average 17 buyer contacts; "selling teams for closed-won deals are 67% larger than those for lost deals" - all **vendor-data**, verified as Gong's published claims, correlational.
- Gong (2020, n=9,056 opportunities): deals without a decision-maker (VP/CXO+) involved are 80% less likely to close; 233% for enterprise deals - **vendor-data**, relative likelihoods, not absolute win-rate points.
- Ebsta x Pavilion (2024, 4.2M opportunities): top performers are far more likely (489% in one section, 241% in a differently-scoped section of the same report) to engage the economic buyer before solution-presented; won deals average 9 engaged contacts at that stage vs 2 for lost - **vendor-data**, platform-customer selection bias.
- LinkedIn Sales Solutions (2016 internal analysis): multithreading lifts win rate 14% overall, 34% for new-business reps; deal sizes 43% lower when single-threaded - **vendor-data**, dated.
- SetSail (2021): 6+ active contacts correlated with a 62% win rate; finance contacts closed best (56%) - **vendor-data**, small vendor, incentive-program context.
- Contact-count heuristics used in the coverage check (~4-7 mid-market, ~10 for $50K-$250K, 11-17+ enterprise) are a convergence of the above plus practitioner guidance - directional, never a quota.
## The Challenger counter-current
CEB's consensus data: purchase likelihood falls from 81% with one decision maker to 55% with two, to the mid-30s at six - **verified** as CEB's published claim (_The Challenger Customer_). Gartner (2025 survey, n=632) found 74% of buying teams show "unhealthy conflict," and consensus-reaching groups are 2.5x more likely to report a high-quality deal - **verified** as Gartner's figures.
Reconciliation with the multithreading data: more stakeholders makes _buying_ harder; more seller relationships across those stakeholders correlates with winning. Multithreading manages buying-group complexity - it does not cause wins, and adding contacts for the count's sake adds consensus burden.
## Champion risk
- ~20% of B2B contacts change jobs annually - **vendor-data** (a job-change-alerting vendor citing a professional-network analysis). The same vendor class reports only ~6% of contacts ever tell a vendor they moved.
- A former champion landing at a new in-profile account is a high-value warm signal (vendor-reported lifts to close rate and deal size are **vendor-data**, promotional context).
- Countermeasures with practitioner consensus: champion redundancy, an executive sponsor whose commitment survives mid-level departures, and a shared plan document acting as institutional memory when a stakeholder inherits the evaluation - **second-hand**.
## Framework disagreements on the roles
- Miller Heiman's "Coach" is not MEDDICC's "Champion": the Coach is a guide who wants you to win and can be low-power; the MEDDICC Champion requires power and influence by definition - **second-hand** comparison, consistent across methodology publishers. Grading by evidence tier resolves the conflict: a low-power guide simply never reaches Tier-1-under-cost proof.
- Challenger rejects title/role-based mapping outright: the Mobilizer is defined by consensus-building disposition, and CEB explicitly contrasts this with Miller Heiman's role-based buying-influence mapping - **second-hand**. This skill sides with behaviour over title, which is why title inference is the floor tier.
- MEDDICC treats qualification (including champion validation) as continuous re-testing through close, versus BANT-style one-time gating - **verified** as MEDDICC's stance; the map's refresh triggers encode the continuous version.
- 30MPC's sequencing frames ("Top-Down Yo-Yo": power -> champion -> blockers; "Bottom-Up Champion Selling": champion -> power -> obstacles) are **verified** as 30MPC's named playbooks - useful when deciding which map gap to attack first. Its stated premise: "a deal without an above-the-line decision maker is a dead deal" (**verified** quote, Armand Farrokh).
## Folklore register (cite only with the label)
- "40% of supposed champions fail the champion test" - practitioner-blog figure, no traceable methodology.
- "79% of purchases require CFO approval"; "C-suite holds final veto in 22% of cases" - restated in aggregator syntheses with no traceable original.
- MAP/deal-room win-rate lifts (26%, 3x, 59/71/45%) - **vendor-data**, self-reported, engagement selects for deals already trending to win.
- CFO-involvement price thresholds (~$100K+) and "security engages whenever the product touches customer data regardless of price" - practitioner heuristics, not a sourced schedule; the second one is conservative and cheap to assume, which is why the workflow assumes it.
references/stakeholder-map-template.md›
# Stakeholder map template
Fill every field or write the honest gap ("unknown - next action X"). An empty cell is a mapping failure; a stated unknown is deal intelligence.
```
STAKEHOLDER MAP
Account : <name> Deal size/segment : <$ / transactional|mid-market|enterprise>
Stage : <stage> Date : <date> Version : <n>
Methodology : <team's methodology, if any - use its vocabulary in the role column>
STAKEHOLDERS (one row per person, present on calls or not)
| Name | Title | Assigned role | Evidence (tier: what they did/said) | Confidence | Gap unproven | Next action |
|------|-------|---------------|-------------------------------------|------------|--------------|-------------|
| | | | | | | |
CHAMPION TEST LOG
Candidate : <name>
Asks made : <capital-costing asks + rung + dates>
Follow-through : <what actually happened, by when>
Next ask : <rung 1-4> - <the ask, with its deadline>
Why this rung : <rung below passed / deleted by RFP or stage / promoting condition met>
Verdict : champion (Tier-1 proven) | coach (hedged/deferred twice) | untested
ECONOMIC BUYER
Named : <name + evidence tier> | UNKNOWN
If unknown : surfacing action - <which role-identifying question, asked of whom>
If access deferred: reading -> countermove <from the deferral table; if no reading is
evidenced, run them in that table's order, top row first>
BLOCKER HYPOTHESES (>=1; >=1 for a function not yet on any call)
Rows in the taxonomy's efficiency order; delete the functions this deal has no seat for.
| Function | Person (or "unidentified") | Signal observed | Veto power | Neutralise/co-opt plan | Effort |
|----------|---------------------------|-----------------|------------|------------------------|--------|
| | | | | | |
COVERAGE
Engaged contacts : <n> vs benchmark <~4-7 mid-market / ~10 / 11-17+ enterprise>
Real relationships (not just email addresses): <n>
Single-threaded : yes/no -> if yes, expansion target: <function + why>
CHAMPION RISK
Redundancy candidate : <name being cultivated> | none (gap)
Executive sponsor : <name> | none (gap)
Job-change detection : alerting in place | manual | none
REFRESH TRIGGER
Re-walk this map on: stage change | new stakeholder appears | champion job-change
signal | <deal-specific event> - whichever comes first.
```
## B2C household adaptation
Same table, different rows:
- Roles column uses the household DMU vocabulary: initiator, influencer, decider, buyer, user, gatekeeper - plus champion (the selling spouse/partner) and detractor (the skeptical relative or co-signer).
- Delete the blocker functions with no household equivalent: procurement, legal, security, compliance. Keep economic (household budget), political (competing family priority), and incumbent loyalist ("the current kitchen is fine").
- Delete the job-change-detection row; champion risk becomes "what happens if the enthusiastic partner cools off before the decider engages?"
- Evidence tiers, the champion test log, EB naming, confidence rules, and the refresh trigger stay exactly as in B2B.
references/worked-examples.md›
# Worked examples
Both examples are this skill's own constructions (**inference**) applying the verified role definitions and the evidence rubric. Rewrite specifics for the actual deal - never reuse verbatim.
## Example 1 - a defensible B2B map
Inputs given: mid-market deal (~$80K ACV) for a workflow-automation product at Meridian Freight (450 employees), evaluation stage. Contact list with titles, three call transcripts, six weeks of email activity, CRM contact roles half-filled.
```
STAKEHOLDER MAP - Meridian Freight, $80K mid-market, Evaluation, v2
| Name | Title | Role | Evidence | Conf. | Gap unproven | Next action |
|------|-------|------|----------|-------|--------------|-------------|
| Dana R. | Dir. Ops | Champion | T1: forwarded our one-pager to the COO
unprompted (email thread); shared the internal vendor shortlist (call 2) | High
| Whether she'll present the case in the exec meeting we won't attend
| Ask her to co-build + present the business case at the Nov ops review |
| Marcus T. | COO | Economic buyer (hypothesis) | T3: Dana says "Marcus decides
anything over $50K"; T4: title fits | Medium | Never met; claim uncorroborated
by behaviour | Dana to arrange 20-min intro; forwardable email drafted |
| Priya S. | IT Manager | Technical blocker | T2: joined call 3, asked only
integration + SSO questions | Medium | Whether she holds veto or advises
| Send the architecture + SSO pack this week (near-zero); hold the POC
unless she names an objection the docs can't answer |
| Leo M. | Procurement | Approver (not yet engaged) | T4: function exists at
this size | Low | Everything | Ask Dana who ran paper on the last purchase |
| (unidentified) | Security reviewer | Silent detractor hypothesis | Product
touches customer data -> assume a reviewer exists | Low | Identity | Ask
Priya: "who signs off on data handling for new tools?" |
| Jen K. | Ops analyst | User/cheerleader | T2: replies within minutes,
attends everything | Low-Med | None load-bearing | Keep warm; not a thread |
CHAMPION TEST LOG
Asks made : rung 1 shortlist share (done, wk 2); rung 2 one-pager forward
(done, wk 2); rung 3 COO intro (committed, pending)
Follow-through : forwarded one-pager same day - T1
Next ask : rung 4 - co-build + present the business case at the Nov ops
review; deadline before that review
Why this rung : rungs 1-2 passed and rung 3 is committed, the decision moves in
an exec meeting we won't attend, and a competitor is shortlisted
- the promoting condition for rung 4, which the efficiency order
would otherwise never reach
Verdict : champion - but re-test at proposal stage
ECONOMIC BUYER : Marcus T. - Medium only; intro pending. If Dana defers the
intro a second time -> reading 2 (fear of exposure) suspected; countermove:
hand her a one-page EB-ready case she presents herself.
COVERAGE : 4 engaged (Dana, Priya, Jen, +CFO office cc'd) vs ~4-7 benchmark.
Real relationships: 2 (Dana, Priya). Not single-threaded, but thin above
the line. Expansion target: Marcus (power), Leo (paper).
CHAMPION RISK : redundancy - Priya, if converted from blocker; exec sponsor -
none (gap); detection - manual only (gap).
REFRESH TRIGGER: Marcus meeting happens, or proposal goes out - whichever first.
```
Why this map passes the gate:
- Every row cites its evidence and tier.
- The only High confidence rests on Tier-1 behaviour.
- The EB is named but honestly held at Medium with the proving action attached.
- A blocker hypothesis exists for someone never seen on a call.
- Single-threading is assessed on relationships, not names.
- The next ask names its rung and the condition that promoted it rather than jumping the ladder.
- Both champion-risk gaps are stated instead of papered over.
## Example 2 - the naive counter-example (do not produce this)
Same inputs. Annotations mark each failure.
```
STAKEHOLDER MAP - Meridian Freight
- Marcus T., COO - DECISION MAKER (high confidence) <- title only (Tier 4):
can never exceed Low; no behaviour, no claim, never met
- Dana R., Dir. Ops - CHAMPION (high confidence) <- right person, wrong
basis: "she's super responsive and loves the product" is Tier 2 warmth;
no capital-costing ask has been logged, so High is unearned
- Jen K., Ops analyst - KEY INFLUENCER <- responsiveness
mistaken for power; she answers fast because it costs her nothing
- Priya S., IT Manager - SUPPORTER (she attended a call) <- attendance read as
support; her questions were all integration risk - blocker signal ignored
- No blockers identified. Everyone we've met is positive. <- maps only who is
on calls; no hypothesis for security, procurement, or finance; the silent
detractor is unmapped by construction
- Next step: send proposal. <- no per-person next
actions, no EB verification, no coverage count, no champion risk, no
refresh trigger - a snapshot pretending to be a map
```
The tell that distinguishes the two: Example 1 states what it does _not_ know and what will prove it; Example 2 converts optimism and titles into confidence. On identical inputs, the naive map forecasts a clean close; the defensible map predicts exactly where this deal can die (unverified EB, unmapped security review, no exec sponsor).
SKILL.md›
---
name: deal-champion-mapping
description: Builds an evidence-graded stakeholder map for one active deal from titles, call notes, email and meeting activity, and CRM contact roles, naming the likely champion, economic buyer, and blockers, each with a confidence level tied to observed behaviour rather than title. Covers B2B buying committees and high-consideration B2C households. Use whenever the user mentions a champion, economic buyer, power map, deal org chart, multi-threading, single-threaded risk, or "who actually decides", even without the word stakeholder. Do NOT use for qualification scoring (mbfinotti/sales-skills@meddpicc-scorecard) or scanning notes for general risk (mbfinotti/sales-skills@deal-red-flags).
license: MIT
metadata:
author: Maya-Beth Finotti
version: "1.4.11"
---
# Champion Mapping
Turn what is already known about an account - contact list with titles, call notes, email and meeting activity, CRM contact roles - into a stakeholder map. The map:
- Names the likely champion, the economic buyer, and the blockers or detractors.
- Grades every assignment by the strength of its evidence.
- States what is still unproven.
- Attaches the next concrete action that proves or disproves it.
The role definitions come from MEDDICC (champion, economic buyer) and Challenger (Mobilizer, Blocker) as named sources; this skill identifies and tests the people - it never scores the deal against any acronym.
Confidence labels used throughout:
- **verified** - confirmed at a named source.
- **second-hand** - documented only via another publisher.
- **vendor-data** - a vendor's own correlational platform dataset, never independently replicated.
- **inference** - this skill's own reasoning rather than a cited claim.
No vendor correlation here is causal, ever.
## Interview
Ask before mapping. One question per message; offer multiple-choice answers when possible; skip anything the user's data already answers.
- B2B, or high-consideration B2C? (B2C here means considered household purchases - remodeling, solar, private school, financial products - not impulse retail.)
- Deal size and segment? (Transactional / mid-market / enterprise - this sets the coverage benchmark and how much committee to expect.)
- What contact data exists? (Contact list with titles, call notes or transcripts, email/meeting activity, CRM contact-role fields - paste or attach whatever is available.)
- What stage is the deal in? (Early discovery, evaluation, proposal, procurement/paper process - later stages demand more proven roles.)
- Does the team run a sales methodology (MEDDICC/MEDDPICC, Challenger, Miller Heiman, other)? The map will use its vocabulary where it matters.
- Industry or regulatory context? (Regulated buyers add compliance and legal seats to the committee by default.)
- Does the product touch customer data? (If yes, assume a security reviewer exists whether or not one has appeared - **inference** from practitioner consensus that security can veto after everyone else approves.)
- What is already known about the org - reporting lines, prior deals at this account, who bought last time?
- Who is the current primary contact, and why them? (The answer often reveals whether the deal is single-threaded.)
- By what date must the next assignment be proven - close date, board meeting, budget cycle? (A hard date promotes the near-zero, same-week asks and deletes anything whose latency exceeds it, starting with the co-built business case.)
- One-off win, or a compounding asset - is this account also a reference, an expansion base, a logo the team will resell against? (A compounding mandate promotes the slow, high-value plays: the co-built case, an executive sponsor, a cultivated second champion.)
- What is the effort ceiling - how much of the contact's political capital can this deal spend, how many rep hours, and can a refused ask be survived? (A low ceiling, or a contact who has already spent capital once, holds the ladder at its cheap rungs and deletes the proof-of-concept.)
## Workflow
1. Run the Interview; collect the contact data before assigning any role.
2. Inventory the evidence. For each contact, separate three piles: observed behaviour (what they demonstrably did), stated claims (what they said about themselves or the process), and title/org inference. These piles are the evidence tiers below - never mix them.
3. Assign a candidate role to each contact from the taxonomy, grading each assignment with the evidence rubric. Contacts with contradictory or empty evidence get "unknown", not a guess.
4. Test the champion candidate: list which capital-costing asks have been made, what the follow-through was, and which rung of the ask ladder to run next - named as a rung, with the reason it was promoted or held.
5. Name the economic buyer, or explicitly flag "EB unknown" with the discovery action that would surface them. If the champion has deferred EB access, classify the deferral (four readings below) and run its countermove; where the evidence names no reading, run them in the section's order.
6. Hypothesize blockers by function from the taxonomy below, in its efficiency order - explicitly including at least one function not yet on any call, and deleting the functions this deal genuinely has no seat for.
7. Run the coverage check against segment benchmarks; flag single-threading as a coverage gap. (A full red-flag review of the deal notes belongs to the deal-red-flags skill, not here.)
8. Note champion risk: is there a second champion candidate, an executive sponsor, exposure to a job change or reorg?
9. Assemble the map per [references/stakeholder-map-template.md](references/stakeholder-map-template.md); run the Quality gate; iterate until every item passes.
10. Deliver with refresh triggers: the map is a living artifact, re-walked at every stage change, new stakeholder appearance, or champion job-change signal - not a one-time deliverable (**second-hand**, Miller Heiman practitioner discipline via published syntheses).
11. If your harness has persistent memory, memorize the map, the evidence behind each assignment, and the open next actions - the next run starts from proven ground instead of re-deriving it. Otherwise hand the user the map file to keep and re-supply.
## Role taxonomy
One-line definition and the distinguishing test for each. Full framework disagreements (Miller Heiman's Coach vs MEDDICC's Champion, Challenger's title-rejection) in [references/evidence-and-benchmarks.md](references/evidence-and-benchmarks.md).
| Role | Definition | Distinguishing test |
| ------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------- |
| Champion | "A person who has power, influence, and, subsequently, credibility within the customer's organization and who is willing and able to assist you" (**verified**, MEDDICC) | Spends internal political capital: passes a capital-costing ask with follow-through |
| Coach / informant | Helpful, shares process and politics, but lacks power or won't spend capital | Gives information; hedges or defers when asked for an EB introduction |
| Cheerleader | Enthusiastic, takes every meeting, no organisational pull | Cannot move budget or override anyone's preference; enthusiasm is the only evidence |
| Sponsor | Senior, lends name and authority to the deal | Endorses but does not do the internal selling work or build the case |
| Mobilizer (Challenger) | Defined by consensus-building disposition, not title (**verified** as Challenger's taxonomy) | Responds to insight with hard questions and drives internal change |
| Economic buyer | "The power to say yes when others say no, and say no when others say yes" (**verified**, MEDDICC) | Can release discretionary or unbudgeted funds for this purchase |
| Budget holder | Owns a budget line | May not have authority to redirect it to this purchase |
| Signatory | Signs the contract | May be a delegate executing the EB's decision - never conflate the two (**second-hand**, MEDDICC guidance) |
| Approver | Required sign-off in the process (procurement, legal, security) | Gate-keeps the paper process; not the value-based decision-maker |
| Blocker | Objects on risk, competing priorities, or politics | Surfaces objections openly; can be addressed or co-opted |
| Detractor / anti-champion | Actively works to defeat the deal, often silently | Rarely on any call; visible only through second-order effects (stalls, resurfaced objections) |
## Evidence tiers and confidence
Grade every role assignment by the strongest evidence supporting it:
- Behaviour under cost beats behaviour.
- Behaviour beats claims.
- Claims beat titles.
| Tier | Evidence | Example |
| ------------------------ | --------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------- |
| 1 - Behaviour under cost | The contact did something that cost effort or political capital | Arranged the EB meeting; forwarded the business case; defended the deal in a meeting the rep didn't attend |
| 2 - Low-cost behaviour | Observable but cheap actions | Fast replies, meeting attendance, sharing non-sensitive information |
| 3 - Stated claim | What they said, uncorroborated | "I'm the decision maker"; "I'll champion this internally" |
| 4 - Title inference | Role guessed from title, department, seniority | "VP of the function that owns this category, so probably the EB" |
Confidence rules (**inference**, this skill's rubric):
- **High**: at least one Tier-1 item, consistent with the rest of the evidence.
- **Medium**: Tier-2 or Tier-3 evidence corroborated by a second independent item.
- **Low**: a single uncorroborated item, or anything Tier-4.
- Title alone is the weakest tier and never yields more than Low - a title can suggest a hypothesis, never confirm a role.
- Missing or contradictory inputs widen uncertainty: cap confidence at Low and mark the role "unknown" rather than defaulting to the typical org chart.
- Company size bends title inference: at small companies function leaders buy directly; at 1,000+ employees the same title often sits two levels from the budget (**second-hand**, published qualification-prompt heuristic).
## The champion test
A champion is proven by an ask that costs them internal capital, and by follow-through - not by warmth. "No Champion, no deal" (**verified**, MEDDICC). The four canonical asks below are a ladder, not a list, ranked on two axes:
- **Effort** - the contact's own political capital, plus calendar latency and reversibility. A refused ask spends standing they may not get back.
- **Value** - evidence strength about whether this person is genuinely a champion.
- effort: co-built case > EB introduction > criteria share > forwardable email
- value: co-built case > EB introduction > criteria share > forwardable email
- efficiency: criteria share > forwardable email > EB introduction > co-built case
- compliance cost: criteria share (under a formal RFP or public procurement, asking for competing bids can disqualify the bid) > forwardable email == EB introduction == co-built case (genuinely equal at zero: none touches a policy, a contract clause, or a sign-off)
Effort and value rank identically, which is why row order can be neither: the strongest proof is also the ask most likely to be refused. Run the ladder in efficiency order.
| Rung | Ask | Contact's effort | Evidence bought |
| ---- | ------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------- | -------------------------------------------------------------------------------------------- |
| 1 | Share internal evaluation criteria, competing vendors, or budget | An hour, plus a norm quietly broken | Tier 1: they took a real risk for you, and the competitive picture arrives in the same move |
| 2 | Forward a rep-written summary internally, or call a customer reference | Near-zero, same day | Weak Tier 1: their name is now attached to you internally - proves willingness, not standing |
| 3 | Arrange a meeting with the economic buyer or an executive sponsor | A week of calendar latency; visible standing spent | Tier 1, plus the deal's single most diagnostic answer - and the EB meeting itself |
| 4 | Co-build - not just receive - the internal business case, and present it at a meeting the rep does not attend | An exec-review cycle; publicly staked, hard to walk back | The strongest evidence available: they defended the deal in a room you could not enter |
Progress the ladder one rung at a time:
- Default to rung 1 with an untested contact.
- Climb exactly one rung each time the previous one passes with follow-through inside its stated timeframe.
- Never repeat a passed rung: a second cheap ask buys no new evidence, and repeating one is how a map stays at Medium confidence for a whole quarter.
Rung 4 is what the efficiency order starves: highest value, worst ratio, last every round, and a collection that only ever computes efficiency never runs it. Promote it anyway when any of these hold:
- Rung 3 has passed, or is structurally impossible (the EB does not meet vendors).
- The decision has moved into a committee the rep cannot attend.
- A competitor is in the room and the map's only remaining risk is whether this contact defends you there.
Delete, do not demote, an ask the deal's constraints rule out; a ruled-out ask parked at the bottom silently reappears as scope.
- Delete rung 1 on a formal RFP or public-sector deal.
- Delete rung 4 before the evaluation stage: a discovery-stage deal has no committee meeting to present at.
This order is a default, not a law. It shifts with segment, stage, and with who executes it: a rep with existing executive relationships pays far less for rung 3 than a first-time seller into the account.
Re-rank against what is already known about this deal:
- An existing relationship with the EB deletes rung 3 (the introduction is already paid for).
- A contact who has already spent capital once starts at the rung above the one they passed.
- A deal too early for a committee holds at rungs 1-2.
Re-rank against the Interview answers too, and say which answer moved which rung.
The diagnostic behind rung 3: "if your Champion is unwilling or unable to introduce you to the EB, this could be a signal that they aren't a true Champion, and you have work to do" (**verified**, MEDDICC). Nate Nasralla's forward-commitment test secures rung 2's proof live during discovery: "if I write you a short email with the summary that we've been working on, are you willing to forward this along?" (**verified**) - the commitment itself, then the follow-through, are both evidence.
Reading the response: follow-through within the stated timeframe is Tier-1 evidence - upgrade. Hedging or deferral twice on capital-costing asks downgrades the contact from champion to coach on the map; do not rest the deal on their strength. (A circulating "40% of supposed champions fail this test" figure is practitioner folklore with no traceable methodology - illustrative only.)
MEDDICC's published warning flags a contact is not a champion (**verified**):
- Lack of information sharing.
- Resistance to making introductions.
- Absence of internal advocacy.
- Limited organizational access.
- Reluctance during internal pushback.
Re-test through the cycle, not once at qualification.
## Economic buyer discovery
"The Economic Buyer is unlikely to raise their hand and tell you they're the EB. It is your job to identify them" (**verified**, MEDDICC). Distinguish EB from budget holder, signatory, and approver per the taxonomy - the EB is the one who could fund this even unbudgeted. Role-identifying questions (this skill's whole allowance of questions - full discovery sets belong to the sales-discovery-questions skill):
- "Who had to approve the last solution you bought like this?"
- "Who is the budget holder for this type of purchase - and could they redirect budget to it?"
- "If everyone else said yes and one person could still say no, who is that?"
**When the champion defers EB access** - one of the most diagnostically loaded events in a deal. Four standard readings, each with its countermove (**second-hand**, practitioner consensus across MEDDICC-derived sources).
Diagnose the reading from evidence rather than picking it, so the ranking below is deliberately narrow: it decides only which countermove to run first when the evidence does not say which reading applies. Ranking a diagnosis the evidence already named would be false precision.
- efficiency (EB access bought per rep hour and per unit of the contact's capital): EB-ready one-pager > decision-path probe > compelling-event revisit > re-run the champion test
- value (what it settles): re-run the champion test > decision-path probe > EB-ready one-pager > compelling-event revisit
| Reading | Countermove | Effort | Odds it opens the door |
| -------------------------------------------------------------- | --------------------------------------------------------------------------------------- | -------------------------------------------------------- | -------------------------------------------------------------------------------------------------- |
| Fears losing control or being exposed as lacking authority | Value exchange: a short, EB-ready business case that makes them look good presenting it | An hour of rep time; near-zero contact capital | Good - and it is the only play that also helps under readings 1 and 3, which is what puts it first |
| Protecting a relationship, or the real decision sits elsewhere | Give them a reason the EB will want the meeting; probe the actual decision path | An hour, on the next call already booked | Moderate on access, high on information - it returns the map's biggest unknown either way |
| No active deal - no urgency | Revisit the compelling event; if none, the map is not the problem | An hour, and the deal may end on the call | Low on access - but a negative answer saves a quarter of misdirected effort |
| Not a true champion - no standing to secure the meeting | Re-run the champion test one rung up; downgrade to coach if it fails again | A week of latency, and the contact's capital spent again | Low - it settles the map rather than opening the door |
Last resort: multithread to the EB through another path. It ranks last on every axis, and what puts it there is irreversibility rather than hours. Going around the champion can destroy trust and convert the champion into a detractor, the single act on this map that turns an asset into a liability (**second-hand**, consistent across practitioner sources).
Delete it outright from a single-threaded deal instead of holding it in reserve: with one real relationship there is nothing left to sell to when it fails. A give-to-get softens the ask instead: trade something the contact wants for the introduction, stated as a condition up front (**verified**, Jason Bay's pattern).
## Blocker taxonomy
Hypothesize blockers by function, not by who has shown up. The recurring pattern in the literature: deals die to a single silent detractor in security, legal, or finance who was never mapped and never on a call (**second-hand**, synthesis of practitioner sources). Table synthesized from the same research base, rows ordered by efficiency - veto risk removed per unit of work, not by seniority or by how loud the function is.
- effort: technical POC > political realignment > ROI model in their numbers > security pack == compliance pack > procurement, legal and incumbent plays (equal at near-zero: each is one send of material you already own)
- efficiency: legal > security == compliance > procurement > economic > incumbent loyalist > political > technical
- compliance cost: compliance/residency commitments > legal redlines > everything else (near-zero) - a residency option or a bespoke clause conceded to clear one review binds every renewal after it, and is the only play here you cannot quietly reverse
Security and compliance tie on both lines because they are the same job - producing an evidence pack - aimed at a different reviewer: an hour when the pack exists, a quarter when it must be built from scratch.
| Blocker | Signal and motivation | Veto power | Neutralise / co-opt | Effort |
| ---------------------------- | ----------------------------------------------------------------------- | ------------------------------------- | -------------------------------------------------------- | ------------------------------------------------------------------------------- |
| Legal | DPA/MSA redlines; liability, data rights | Blocks on unacceptable clauses | Pre-empt with standard terms | Near-zero: send the standard paper before they ask |
| Security / InfoSec | Questionnaire, certification review; data isolation, compliance posture | Can veto after everyone else approves | Certifications, architecture docs, SSO | An hour if the pack exists; a quarter to build one |
| Compliance / privacy | Data-residency and regulatory review; audit mandates | Veto on residency/compliance | Compliance briefs; residency options | An hour if the pack exists; a quarter to build one |
| Procurement | Appears at paper-process stage; price, terms, vendor risk | Controls contract terms and timeline | Engage early; give champion negotiation cover | Near-zero, but only if run before the paper stage |
| Economic (CFO, budget owner) | TCO/ROI questions; controls funding; cost and opportunity cost | Blocks funding if ROI unproven | ROI model in their own numbers; business case in dollars | A week - and it needs the champion's capital to get their numbers |
| Incumbent loyalist | Advocates status quo; switching cost, sunk relationship | Rallies for "do nothing" | Cost-of-inaction framing; contain rather than convert | Near-zero to produce, but conversion rarely lands - budget for containment only |
| Political | Competing initiative, turf; status, control, rival project | Informal; can rally opposition | Align to their agenda; executive sponsor | A week or more, and unwinnable if their rival project is funded |
| Technical (CTO, architect) | "We need to run this through IT"; integration cleanliness, standards | Veto authority, rarely budget | Surface early; technical validation, POC | A quarter of engineering time on both sides |
The proof-of-concept is what this order starves: it produces the strongest technical proof and has the worst ratio on the table, and it is the only play that also spends the buyer's engineering time and the champion's capital to schedule. Promote it when either holds:
- The technical reviewer holds the real veto and has stated an integration objection the documents cannot answer.
- The product's differentiation is only visible inside the buyer's own environment.
Delete rows this deal rules out rather than carrying them low:
- The compliance/privacy row for an unregulated buyer whose product touches no personal data.
- The whole legal/security/compliance/procurement block for B2C households (see the template's adaptation).
A row parked at the bottom returns as unplanned scope at the paper stage.
Watch the seniority mismatch: mapping over-indexes on the executive (the CISO) while the person who runs the actual review (the security architect) holds informal veto over anything that doesn't integrate cleanly (**second-hand**). The most dangerous detractor is often not on any call - the map must contain hypotheses about absent people, not just observations about present ones. Re-rank the plays against what this account already gives you: an existing security-review pass from a prior deal, an in-house solutions architect, or a compliance pack already written for another buyer all move their row up.
## Coverage check
- Benchmarks by segment (**vendor-data** and practitioner heuristics, correlational, not targets): roughly 4-7 engaged contacts mid-market, ~10 for $50K-$250K, 11-17+ for strategic enterprise. Gartner's survey puts the median buying group at 6-10 decision makers (**verified** as Gartner's figure, 2017, n=750).
- Single-threading is a coverage gap on this map: one real relationship means the deal dies with one departure. Treat any single-threaded deal over $50K as structurally behind - multithreaded deals over $50K win at materially higher rates in Gong's 1.8M-deal analysis (its "130%" lift is **vendor-data**, correlational). Flag the gap here; the deal-red-flags skill owns the broader review.
- Count relationships, not names: "whether you're actually protected in a deal - or just have multiple email addresses and one real relationship" (**verified** quote, 30MPC/Gong report).
- The Challenger counter-current, stated so the map isn't read as "more contacts = better": CEB's data shows purchase likelihood _falls_ as buying-group size grows (81% at one decision maker, 55% at two, mid-30s at six) because consensus gets harder (**verified** as CEB's published claim). More stakeholders makes buying harder; multithreading is how a seller _manages_ buying-group complexity, not a force that causes winning. Add contacts to cover real roles, never to inflate a count.
## Champion risk
- Roughly 20% of B2B contacts change jobs in a year (**vendor-data**, a job-change-alerting vendor citing a professional-network analysis) - a multi-month deal resting on one champion is structurally fragile, and most departures are silent.
- The map must answer three questions:
- Is there a second champion candidate being cultivated?
- Is there an executive sponsor whose commitment survives a mid-level departure?
- Would a job change be detected within days (alerting) or discovered at the next ghosted email?
- On a champion-departure signal: activate the redundancy candidate immediately, lean on the executive sponsor for continuity - and note the departure is also an opportunity at the champion's next company.
## B2B and B2C
High-consideration B2C runs the same multi-role logic under different names: consumer theory's household decision-making unit (**verified**, Schiffman & Kanuk / AMA academic taxonomy).
- **initiator** - raises the need
- **influencer** - shapes criteria
- **decider** - holds authority
- **buyer** - executes the purchase
- **user** - consumes
- **gatekeeper** - controls information flow
Mapping onto this skill's roles (**inference**, analytic transfer, not empirically measured equivalence):
- decider ≈ economic buyer
- gatekeeper ≈ the administrative/procurement screen
- influencer ≈ technical evaluator
- user ≈ end user
- the enthusiastic spouse who sells the other is the champion
- the skeptical parent, in-law, or co-signer is the silent detractor
- "do nothing" is as real a competitor in a kitchen renovation as in enterprise software
Works identically for both:
- The evidence tiers.
- The champion test (an ask that costs the enthusiastic partner effort - "will you walk your spouse through this quote before Thursday?").
- The deferral readings.
- Confidence rules.
- The quality gate.
Say so on the map rather than leaving it implicit.
Does not transfer:
- No procurement, legal, or security veto function, no written paper process, and no CRM tooling.
- Power maps become informal intuition.
- Roles shift per purchase (the decider for the car may be a mere influencer for the school).
- Family emotional dynamics are not captured by influence/sentiment scoring.
Drop those rows from the template; keep everything else.
## Invocation and expected output
Typical invocations:
- "Here's the contact list and my call notes for the Meridian deal - who's my champion, who actually signs off, and who's going to block this?"
- "Map the stakeholders on this opportunity; my main contact is great but I can't tell if she's a champion or just friendly."
- "We sell home solar; the husband booked the consult and loves it - map the household before I send the proposal."
Deliver one stakeholder map (full template in [references/stakeholder-map-template.md](references/stakeholder-map-template.md)):
```
STAKEHOLDER MAP - <account>, <deal size/segment>, <stage>, <date>
Per stakeholder : name, title, assigned role, evidence tier + the evidence itself,
confidence, gap still unproven, next action to prove/disprove
Champion test : asks made, follow-through observed, next ask + its rung and why
Economic buyer : named + evidence, or "unknown" + the surfacing action
Blockers : >=1 hypothesis per plausible function, incl. people not yet on calls
Coverage : engaged-contact count vs segment benchmark; single-thread flag
Champion risk : redundancy candidate, executive sponsor, job-change exposure
Refresh trigger : what event forces the next re-walk of this map
```
## Common failure modes
| Failure | Fix |
| ----------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Mistaking responsiveness for power | Fast replies are Tier 2; the contact who takes every meeting may have zero organizational pull - grade by capital spent, not warmth |
| Assigning champion by title | Title is Tier 4, capped at Low; a champion is proven by a passed capital-costing ask, nothing less |
| Mapping only who is on calls | The most dangerous detractor never joins one; force blocker hypotheses for absent functions |
| Treating the map as a one-time artifact | Re-walk at every stage change, new stakeholder, or job-change signal; unmet roles and unreadable stances are the most valuable items on it |
| Going around the champion to reach the EB | Run the deferral countermoves in their ranked order first; the bypass converts champions into detractors, and is deleted outright on a single-threaded deal |
| Reaching for the biggest ask, or repeating the cheapest one | Climb the ask ladder one rung at a time from rung 1; a rung already passed buys no new evidence, and rung 4 needs its promoting condition stated |
| Conflating EB with signatory or budget holder | The signatory may be a delegate; the budget holder may lack redirect authority; test for "yes when others say no" |
| Counting email addresses as coverage | Multiple contacted names with one real relationship is still single-threaded |
| Reading a stated claim as proof | "I'm the decision maker" is Tier 3; corroborate with behaviour before raising confidence past Medium |
## Quality gate
Score the drafted map against all eight. Pass threshold: 8/8. Iterate until nothing fails.
1. Every named stakeholder carries a role, an evidence tier with the actual evidence cited, a confidence level, and a next action.
2. No High-confidence assignment rests on title alone; no champion is High-confidence without a passed capital-costing test, and the next ask names its rung plus why that rung and not the one above.
3. The economic buyer is named with evidence, or explicitly flagged unknown with the surfacing action - never silently omitted.
4. At least one blocker/detractor hypothesis exists, and at least one names a function not yet present on any call.
5. The coverage count is compared to the segment benchmark and single-threading is flagged if present.
6. Champion risk is answered: redundancy candidate, executive sponsor, and detection method each addressed (or honestly marked absent).
7. Every vendor statistic on the map is labeled correlational; no stated claim is presented as proven behaviour.
8. A refresh trigger is stated - the map declares when it goes stale.
## KPIs and measurement
- Per refresh, the map worked if:
- Next actions from the prior version were executed and their assignments proved or disproved (an assignment that never moves off Tier 3-4 across two refreshes is a gap, not a fact).
- The champion passed a new capital-costing ask.
- No stakeholder appeared at proposal stage who wasn't at least hypothesized earlier.
- Across deals, track the share of maps where the EB was engaged before the proposal/solution-presented stage - top performers engage the EB early far more often in Ebsta/Pavilion's multi-million-opportunity analyses (**vendor-data**, correlational, platform-selection bias applies) - and the share of losses attributable to an unmapped detractor. Falling unmapped-detractor losses is the honest success signal.
Optional integration note:
- CRM contact-role fields seed Tier-4 hypotheses.
- Conversation-intelligence transcripts supply the Tier-1/Tier-2 behavioural evidence.
- Job-change alerting feeds the champion-risk row.
All three are categories - any tool of the class works, and the map functions with none of them.
## Reference
- [references/stakeholder-map-template.md](references/stakeholder-map-template.md) - fill-in-ready map template with B2C adaptations
- [references/worked-examples.md](references/worked-examples.md) - complete worked B2B map and an annotated naive counter-example
- [references/evidence-and-benchmarks.md](references/evidence-and-benchmarks.md) - buying-committee size figures, multithreading data with caveats, framework disagreements on role definitions, misquotation warnings
- `mbfinotti/sales-skills@meddpicc-scorecard` for scoring the deal against MEDDPICC criteria
- `mbfinotti/sales-skills@sales-discovery-questions` for building the discovery question set
- `mbfinotti/sales-skills@deal-red-flags` for the general red-flag review of deal notes