Section 1
Key takeaways
• 92% of B2B buying decisions are made by groups of two or more people, so assuming a single decision-maker, even on a small engagement, is the wrong default. • In small and mid-size deals the committee is real but small: in one 2026 survey, 50% of buying groups had just 2-4 people and 42% had 5-9, and every group of 10+ came from a company over 1,000 employees . • More than 80% of sellers say a deal stalled or was lost in the past year because a key stakeholder left, proof that the unmapped person, not the competitor, is what kills small-company deals. • 79% of purchases require CFO approval , so the finance seat is almost always occupied even when the founder never meets it. • Buyers spend only 17% of their journey with suppliers and 5-6% with any one rep, you get a sliver of time, so you have to map the committee fast.
Section 2
Why founders are sure there's only one decision-maker (and why they're wrong)
The single-decision-maker belief feels true because of how small deals present themselves. One person books the call. One person nods through the demo. One person says "this looks great, let me get it moving." From the founder's chair, that looks like the whole sale. It isn't. It's the visible 17%. Gartner found that B2B buyers spend only 17% of their entire purchase journey meeting with potential suppliers, and just 5-6% of that time with any single sales rep . The other 83% happens in hallways, Slack threads, and internal forwards you never see. The person on your call is the tip; the committee is the rest of the iceberg, and it's making decisions in rooms you're not in. And the data on whether those rooms are crowded is unambiguous. 92% of B2B buying decisions are now made by groups of two or more people . That figure includes the small stuff, it isn't an enterprise-only phenomenon. Once you internalize it, "I just need to convince one person" stops sounding like efficiency and starts sounding like a missing map. There's a counter-argument worth taking seriously: in a genuinely small company, isn't the owner often both decider and budget? Sometimes, yes. But "one person holds two seats" is not the same as "there is only one seat." The owner who signs the check still has an operations lead who has to live with your service every day, and that lead can quietly starve adoption until the renewal conversation becomes a cancellation conversation. The seats exist whether or not different bodies fill them. Your job in discovery is to count the seats, not the bodies.
Section 3
How big is the committee in a small company, really?
This is where the popular statistics mislead founders in the opposite direction. You've seen the headlines: the average B2B buying group has six to ten people, sometimes more . Those numbers are real, but they're enterprise averages, and applying them to a five-person engagement makes the problem feel so big you don't bother mapping it at all. The grounded number is smaller. In Influ2's 2026 survey of 50 enterprise and mid-market buyers, 50% said their buying group had just 2-4 people and 42% said 5-9, and every respondent with a group of 10 or more came from a company over 1,000 employees . Read that again, because it reframes the whole task: in small and mid-size deals, the committee is real but small. You're not mapping a parliament. You're mapping two to five people. That is genuinely good news, and it's the practical heart of this piece. A 13-person enterprise committee is a research project. A four-person committee is a discovery call. You can hold four roles in your head. You can tailor four messages. The reason founders don't is not that the task is too big, it's that they don't believe the other three seats exist until the deal dies. The committee has also been growing, slowly, for a decade, which is why even seasoned operators are calibrated to an older, simpler buying motion. As Brent Adamson and his co-authors wrote in Harvard Business Review back in 2017: "The number of people involved in B2B solutions purchases has climbed from an average of 5.4 two years ago to 6.8 today" . That drift has continued. If your selling instincts were formed when one champion could carry a deal, your instincts are now a few stakeholders short.
Section 4
The stakeholder you never met is the one who kills the deal
Here is the most expensive pattern in small-company sales, stated plainly: the deal doesn't die because someone said no. It dies because someone you never identified changed the math while you weren't looking. The clearest evidence is about stakeholder movement. More than 80% of sellers say a deal stalled or was lost in the past 12 months because a key stakeholder left . Think about what that requires to be true. It requires that the deal depended on a stakeholder the seller knew mattered, which means in the deals that died, there was a person holding the outcome who could vanish and take it with them. It also means the deal was single-threaded, hung on one relationship, which is a documented way to lose winnable deals even before you account for the stakeholders you never mapped at all. The hidden blocker and the silent decider aren't edge cases. They are, statistically, the modal cause of a lost deal. The finance seat is the most consistently invisible of the four. 79% of purchases require CFO approval, and 52% of buying groups now include a decision-maker at VP level or above . In a small-company deal, you almost never meet that person. Your champion mentions "I just need to run it by finance" as if it were a formality, and you treat it as one. It is not a formality. It is a separate buyer, with a separate question, who will evaluate your service on a dimension your champion never raised, and who can kill a deal everyone on the call loved. This is the same failure mode that quietly sinks demos and proposals: you sell to the enthusiasm in the room and lose to the silence outside it. If you want the demo-stage version of this problem, handling the objection from the person who isn't there, that lives in the playbook for surfacing and answering hidden objections before they harden. For now, hold the principle: the named risk is survivable; the unnamed one is fatal.
Section 5
The four seats every deal has
Strip away company size and industry and every B2B purchase routes through four functional jobs. Different people may fill them, one person may hold two, but the jobs are always present. The Decider owns the outcome. This is the person whose problem the purchase solves and whose judgment carries the room. They care about results, not features. In a small company this is often the owner or a department head, and critically, it is frequently not the person who booked your call. The User lives in your service every day after you're hired. The ops manager who'll run your reporting, the marketing coordinator who'll use your system, the assistant who'll field your deliverables. The User rarely has formal veto power, but they have something more dangerous: the ability to not adopt. A User who feels threatened or burdened will comply on paper and resist in practice until the engagement looks like a failure. Finance owns the risk. Sometimes a literal CFO, sometimes the owner wearing the money hat, sometimes a controller you'll never meet. Finance doesn't evaluate whether your service is good. Finance evaluates whether the spend is defensible if it goes wrong. Different question, different language. The Blocker is whoever has a reason to prefer the status quo. The IT person who doesn't want another integration. The incumbent vendor's internal advocate. The team lead who proposed the in-house alternative. Blockers are not villains, they usually have a legitimate concern, but a Blocker whose concern is never surfaced becomes the anonymous "let's hold off" that ends the deal. Notice these map onto the roles the sales literature has long named, champion, end-user, evaluator, blocker, economic buyer, but compressed to the four that actually move a small deal. You don't need a seven-box stakeholder grid for a $20k engagement. You need to know who owns the outcome, who lives in it, who owns the money, and who's quietly rooting against you.
Section 6
The BGA framework: The D-U-F-B Map (Decider / User / Finance / Blocker)
Every deal, regardless of size, has four functional seats, Decider, User, Finance, Blocker, even when one person fills two of them. The D-U-F-B Map is how you surface all four during discovery and then sell to each one in its own language. It is the discovery-side companion to qualifying the opportunity itself, which is covered in the qualification framework for deciding whether a deal is even real. Step 1, Surface the Decider and Finance with the "how a yes happens" question. Ask, in these words or close to them: "Walk me through how a yes actually happens here, who signs, and who has to be okay with it before it does?" This single question does double duty. "Who signs" surfaces Finance. "Who has to be okay with it" surfaces the Decider if it isn't the person in front of you. Listen for the throwaway clause, "well, I'd loop in Dana", and treat every name as a seat to be mapped, not a detail to be ignored. Metric: if you end discovery and cannot name the person who signs, your map is incomplete and your forecast is fiction. Step 2, Surface the User with the "who lives in this" question. Ask: "Once we're live, who's actually using this every day?" The answer is your User, and it is almost never the same person as the Decider. Then ask the User-specific follow-up: "What does their current day look like, and what would change?" You're hunting for the workflow your service threatens, because that threat is the seed of non-adoption. Metric: you should be able to name at least one concrete thing that gets easier for the User. If you can't, you don't have a User-side message yet. Step 3, Surface the Blocker with the "what stalls this" question. Ask, directly: "If this stalls three months from now, who's most likely to be the reason, and why?" Good prospects will tell you. They'll name the IT constraint, the budget-cycle timing, the colleague who likes the current vendor. You've just been handed the objection before it became fatal. Metric: you should leave discovery with a named Blocker and their specific concern. "There's no blocker" is not an answer; it's a flag that you haven't found them yet. Step 4, Tailor one message per seat. This is the step founders skip, and it's the one that wins deals. The four seats buy four different things: • The Decider buys the outcome. Lead with the result, the before-and-after, the business case. Spare them the workflow detail. • Finance buys the risk-adjusted return. Not "this will grow revenue" but "here's the cost, here's the realistic range of outcomes, here's what happens if it underperforms, and here's why the downside is contained." Finance is reassured by your willingness to name the downside, not by your optimism. • The User buys "this makes my day easier, not harder." Show them the specific friction you remove. A User who believes your service makes them more valuable becomes your strongest internal advocate; one who fears it makes them redundant becomes your quietest Blocker. • The Blocker is neutralized by addressing their objection before they raise it. If you know IT worries about integration load, bring the integration answer to the table unprompted. A surfaced objection is a conversation. An unsurfaced one is a no with no return address. Step 5, Arm your champion to sell when you're not in the room. Remember the 17%: most of the decision happens without you . Your champion is your proxy in the other 83%, but only if you've equipped them with the seat-specific messages that belong in a champion's toolkit, the one-line ROI framing for Finance, the day-easier story for the User, the integration answer for the Blocker. A champion armed with one outcome pitch can only sell the Decider. A champion armed with four messages can sell the room. Once the committee is mapped and messaged, keeping it warm through the decision window is a follow-up systems problem, handled in the cadence playbook for staying present across a multi-week buying cycle. The founder's most common, most expensive mistake is selling the whole committee the same outcome-pitch that only the Decider cares about, and never naming the Blocker until the deal is already dead. The D-U-F-B Map exists to make both of those impossible. A worked example A founder selling a $2,500/month fractional-ops service gets a warm call from a startup's head of marketing. The marketing head is excited, fast, easy, "let's do it." The untrained instinct is to send the proposal and wait. Run D-U-F-B instead. The "how a yes happens" question reveals the founder, not the marketing head, signs anything over $2k, Decider and Finance are the same person, and they're not on the call. The "who lives in this" question reveals an operations coordinator who currently owns the workflows this service would absorb, the User, and a potential Blocker if she reads it as her job being outsourced. The "what stalls this" question gets the marketing head to admit: "Honestly, if anyone slows this down it's Priya, she built our current process." Now the founder sells four messages. To the marketing head (champion), the outcome story plus the tools to carry it up. To the absent founder (Decider/Finance), a one-page risk-framed case: cost, expected return, and what a 90-day pilot de-risks. To Priya (User/Blocker), a direct conversation positioning the service as removing her lowest-value work so she can own higher-value projects, turning the likely Blocker into a User who benefits. The deal that would have died in the founder's inbox closes, because the founder mapped the three seats the champion never mentioned. That's the whole discipline. The committee was always there. The map just made it visible in time to do something about it. If you want the three discovery questions and the per-seat message map as a fill-in worksheet you can run on your next call, the Template Pack includes a D-U-F-B discovery template, and the deeper mechanics live in the LeadOS playbook.
Section 7
You're running the D-U-F-B Map right when…
You're running the D-U-F-B Map right when you can end every discovery call able to name a specific person, or an explicit "one person holds this seat", for all four functions, and you've left with a named Blocker and their actual concern rather than a comfortable "no objections." You're running it right when your proposal isn't one document but one core case plus three tailored angles: an outcome story for the Decider, a risk-framed return for Finance, a day-easier story for the User. You're running it right when your champion can sell the deal in a room you're not in, because you handed them the language for every seat. And the clearest signal of all: you stop being surprised. When a deal stalls, you already know which seat moved, because you mapped it before it could become the anonymous person who killed a deal you should have won.