Section 1
Key takeaways
• A single warm contact represents roughly 8% of a typical B2B buying decision, Forrester puts the average purchase at 13 stakeholders , so betting the forecast on one person is a structural error, not a judgment call. • Deals don't usually die to competitors. 86% of B2B purchases stall internally , which means your real opponent is the committee's inability to align, not the rival on the shortlist. • The lever is consensus: buying groups that reach agreement are 2.5 times more likely to call their deal high-quality . The seller's job is to engineer that consensus across roles, not to charm one person. • Five roles exist in almost every purchase, economic buyer, decider, influencer, blocker, champion, even inside a ten-person company. Map roles, not headcount. • The fastest way to lose is the Committee-of-One Trap: treating one enthusiastic contact as the whole deal and discovering the blocker only when they say no.
Section 2
Why a single contact is structurally incomplete
Start with the arithmetic, because it ends the debate fast. Forrester's State of Business Buying puts the average B2B purchase at 13 stakeholders . If you have one engaged contact, you are looking at roughly 8% of the people who can move or kill your deal. You wouldn't underwrite a loan on 8% of the financials. You're doing the equivalent every time you forecast off one champion. People assume this is an enterprise problem, that committees are a Fortune 500 phenomenon and a thirty-person agency buys the way a freelancer does. The data says otherwise. Buying committees have grown from 5.4 stakeholders in 2015 to 8 to 13 in 2025, nearly doubling in a single decade . "Small company, single buyer" is now the exception, not the rule. Even a mid-market deal pulls in an owner, an operations lead, whoever owns the budget line, and usually one skeptic from finance or IT. And these people don't show up aligned. For complex solutions, Gartner's benchmark is a buying group of 6 to 10 decision makers, each arriving with their own independent research, typically 4 to 5 pieces, that they then reconcile against each other . Picture that mechanically: six people walk in with six different mental models of the problem, the budget, and the alternatives, and the purchase only happens once they grind those down to something they can all live with. That reconciliation is the 74% conflict Gartner measures . It's not a personality problem. It's structural friction built into group buying, and most of it happens in rooms you're not in. As Delainey Kirkwood, a principal in the Gartner Sales Practice, puts it: "Buying groups are more diverse than ever, ranging from five to 16 people across as many as four functions" . Four functions means four sets of incentives. The marketing lead wants speed, finance wants the lowest total cost, operations wants no disruption, and IT wants nothing new to maintain. A pitch that delights one of them can actively threaten another, which is exactly how a deal with a thrilled champion still dies. This is the failure that the rest of your sales motion can't recover from. You can have flawless positioning that frames the problem before you ever pitch and a discovery process that qualifies hard, and still lose because you mapped one person instead of the room. Mapping isn't a nicety layered on top of good selling. It's the load-bearing wall.
Section 3
What the buying committee actually looks like (even when it's three people)
Here's the reframe that makes this usable for a service business: stop counting heads and start counting roles. A ten-person company doesn't have a thirteen-person committee, but it almost always has all five roles, sometimes with two roles in one person, sometimes with a role that's invisible until it stops you. Map the roles, and the headcount sorts itself out. Five roles run nearly every purchase: The Economic Buyer controls the budget and signs off on the final spend . They care about return, risk, and opportunity cost, not your features. In a small firm this is often the owner or a partner. The economic buyer can be completely absent from every call until the number gets real, then materialize and reset the entire conversation. An unmapped economic buyer is the single most common reason a "closed-won" forecast turns into a "stalled" one. The Decider holds final approval authority on the solution itself . Sometimes this is the same person as the economic buyer; often it isn't. A department head might decide which vendor, while the owner decides whether there's budget at all. Confusing these two is how you spend three weeks selling hard to someone who can say no but cannot say yes. The Influencer shapes opinion but owns no decision . They have the ear of the decider, a trusted senior employee, an outside advisor, the founder's spouse who happens to run a similar business. Influencers are easy to dismiss because they have no title in the deal, which is precisely why they're dangerous. They lobby in private, and you usually learn they existed only after they've already moved the decider. The Blocker can slow or kill the deal from a function you may never pitch to, legal, IT, procurement, security, compliance. Blockers rarely want to buy; their job is to find reasons not to, or conditions that must be met first. A blocker isn't an enemy. They're a gate, and gates have keys. But you can't find the key to a gate you didn't know existed. The Champion is your internal advocate, the person who sells for you when you're not in the room . They want this to happen, they'll spend their own credibility to push it, and they'll tell you what's really going on internally if you ask. A real champion is worth more than any feature. But, and this is the whole point of the article, a champion is necessary and nowhere near sufficient. A champion with no map of their own committee is just an enthusiastic person who can't close. You'll see longer taxonomies in the wild, end users, technical buyers, executive sponsors, ratifiers . Use them when they apply. But these five are the irreducible set. If you can name a real person for each role, or consciously confirm a role is absent, you have a map. If you can't, you have a contact and a hope.
Section 4
The Committee-of-One Trap
Let me name the specific failure mode so you can catch yourself in it, because it's seductive and it feels like momentum. A bookkeeping firm gets a warm inbound from the operations manager at a forty-person construction company. Great call. The ops manager hates their current bookkeeper, loves the firm's monthly-close process, and asks for a proposal. The firm sends it, the ops manager says "this looks perfect, let me run it up," and the deal goes quiet for six weeks before dying with a vague "we decided to hold off." What actually happened: the ops manager was an influencer, not a decider. The owner, the economic buyer, never saw the value framed in terms he cared about, only a price. And the company's outside CPA, an influencer with the owner's ear, said "you don't need to switch, I can handle the cleanup." The bookkeeping firm sold beautifully to one role on the committee and never knew the rest of the room existed. That's the Committee-of-One Trap: treating one warm contact as the deal, mistaking enthusiasm for authority, and discovering the rest of the room only when it says no. The trap is dangerous precisely because the warm contact gives you everything that feels like progress, engagement, fast replies, positive language, while hiding the variable that actually decides the outcome. You don't feel under-informed. You feel like you're winning. The forecast goes up. And the contact, who genuinely likes you, has no incentive and often no ability to tell you they can't get it done. They may not even know themselves. The antidote is not working the contact harder. It's mapping the room around them, turning your one relationship into a question-asking instrument instead of a closing target.
Section 5
How do you map roles during discovery?
You map during discovery, with questions, before you've earned the right to a proposal. Mapping is not a separate phase; it's a lens you hold over every conversation. Three questions do most of the work, and they're disarmingly simple because the goal is to get named people, not vague org-chart theory. Ask "Besides you, who else weighs in on a decision like this?" This surfaces the committee without making your contact defensive. You're not asking "are you the decision maker?", a question that wounds ego and invites a lie. You're assuming a group, which is accurate, and letting them populate it. Ask "Whose budget does this come out of?" This finds the economic buyer directly. The answer is often a different person than your contact, and the gap between "who I'm talking to" and "whose money it is" is where most stalls live. Ask "Who, if anyone, tends to push back on changes like this?" This finds the blocker before the blocker finds you. People will happily name the skeptic in their org because it doesn't feel disloyal, it feels like helpful context. That's your IT lead, your procurement gate, your cautious CFO, named out loud, weeks before they could have ambushed you. The discipline that makes this real: for every name you hear, assign a role and a status. Status has three values, mapped (you know they exist and their role), engaged (you've had real contact and read their position), and unaddressed (you know they matter and you've done nothing about it). The unaddressed column is your actual pipeline risk, and it's invisible on a normal CRM stage. A deal sitting in "proposal sent" with an unaddressed economic buyer and an unaddressed blocker isn't late-stage. It's unmapped, and it will teach you that the hard way. This is also where mapping connects to the rest of the deal mechanics. The roles you surface here determine which objections you'll face later, the blocker's concerns are not the champion's concerns, which is why a clean map feeds directly into the objection-handling work that happens at the demo and close. Map early, and you stop being surprised. For the question scripts and a fill-in stakeholder grid, the template pack has the mapping worksheet ready to use.
Section 6
The BGA framework: Role-Not-Headcount Mapping
Here's the system. It's built to work in a service business with a normal CRM and no special tooling, five steps, run during discovery and updated every time you learn something new. 1. Assume the committee, then size it. Default to a group, never a single buyer. On the first real call, ask "Besides you, who else weighs in on a decision like this?" and write down every name. Rule of thumb: if you've heard fewer than three names by the end of discovery on a deal of any real size, you haven't mapped, you've talked to one person. Forrester's average is 13 stakeholders ; you won't engage all of them, but you should be able to name more than one. 2. Assign every name a role. For each person, tag exactly one primary role: economic buyer, decider, influencer, blocker, champion. Force the choice, "they're kind of both" usually means you don't actually know, which is itself useful information. The two you most often miss are the economic buyer (because they hide until the number is real) and the blocker (because they live in a function you never pitch to). If those two cells are empty, that's your next two questions, not your next proposal. 3. Set a status on every role. Mark each named person mapped, engaged, or unaddressed. Then look only at the unaddressed column, that's your risk register for this deal. The goal across the cycle is to move the economic buyer, the decider, and any blocker out of "unaddressed" before you treat the deal as real pipeline. 4. Engineer consensus, don't just collect contacts. Mapping is the means; alignment is the end. Gartner's own data is blunt here: buying groups that reach consensus are 2.5 times more likely to call the deal high-quality . So your job after mapping is to give each role what they need to say yes, return framing for the economic buyer, a clean answer for the blocker, ammunition for the champion to use when you're not there. Equip your champion to sell internally; they spend more time in those private rooms than you ever will, and that's where the 74% conflict gets resolved . 5. Re-map on every new signal. A buying-process map is a living document, not a one-time form. New name on a CC line? Map them. Your champion changes roles? Re-map. The deal goes quiet? Check which role went silent, silence from the economic buyer means something very different than silence from an influencer. Update the grid every time the picture changes, and the map will warn you about a stall weeks before the CRM stage does. This is the discipline that later hands clean, current account intelligence to your follow-up and renewal systems instead of a stale contact record. The whole framework fits on one screen: a five-row grid, role, named person, status, that you keep beside the deal. It costs you three questions and the honesty to leave cells blank when you don't know. What it buys you is the end of the surprise stall.
Section 7
You're running Role-Not-Headcount Mapping right when…
You're running it right when you can pull up any open deal and name a real person, or a confirmed absence, for all five roles, with a status next to each. When "the deal is going great" is a sentence you no longer trust on its own, because you've learned to ask "great according to which role?" When a quiet week makes you check which role went silent rather than just sending a nudge to your one contact. When your champion has a one-page map of their own committee that you built together, and they're using your language to sell finance and IT in rooms you'll never enter. And when an unaddressed economic buyer or blocker stops a deal from advancing in your own mind, not because a rule says so, but because you've internalized that an unmapped deal is a guess wearing the costume of a forecast. You're running it wrong when your pipeline is a list of warm contacts, each one a Committee-of-One, every forecast resting on someone who likes you but may not be able to help you.