Buying groups: why your champion is outnumbered and how to sell to consensus

Most sales methods still center on the champion: the person inside the buyer who wants the change. The champion matters. But the champion is now one voice in a crowded room, and most of that room never meets the seller. This guide gives the current numbers on buying groups, explains why consensus predicts deal quality, and sets out a practical kit for selling to the whole group.

How big is a B2B buying group in 2026?

A typical B2B buying group has about ten or more people, and recent research puts it higher once outside advisers are counted. The main public sources:

Definitions differ between studies, so do not compare the numbers too closely. The direction is consistent. Purchases now cross departments, and buyers consult people outside the company, such as peers, advisers and analysts.

Why is your champion outnumbered?

Your champion is outnumbered because every stakeholder forms a view of every vendor, and the champion must reconcile all of them. Do the arithmetic. Thirteen internal stakeholders and five vendors make 65 stakeholder-vendor pairs. That is our arithmetic, not a research finding. But it shows the load on one person. Most of those 65 views form in meetings you are not in.

Conflict makes it harder. A Gartner survey of 632 B2B buyers, published in May 2025, found that 74% of buying teams showed unhealthy conflict during the decision. Gartner defines unhealthy conflict as members having conflicting objectives, disagreeing on the best course of action, or being overruled by outside decision-makers. Forrester reported in 2024 that 86% of B2B purchases stall during the buying process.

The disagreement is often about the problem, not the vendor. Finance sees a cost issue. Operations sees a process issue. IT sees a risk. If the group never agrees on what the problem is, no vendor comparison can settle it.

Why does consensus matter more than the champion?

Consensus matters more because it predicts whether buyers judge the deal a good one. In the same Gartner study, buying groups that reached consensus were 2.5 times more likely to report a high-quality deal. Buyers who experienced buying-group relevance were three times more likely to report a high-quality deal.

Gartner also found a trap. Content tailored to each individual's priorities had a 59% negative impact on buying-group consensus. Tailoring for the group's shared relevance improved consensus by 20%. Personalization that pleases each stakeholder alone can push them apart. The aim is a shared story, with each role's concern answered inside it.

How do you sell to consensus?

Sell to consensus by getting the group to agree on the problem and its cost before you compare solutions. We use a six-part consensus kit:

1. One problem statement, in their words. "Your stage-2 drop-off", not "pipeline challenges". It should fit in three sentences.
2. A cost the CFO recognizes. Put the problem in money per quarter, using the buyer's own numbers.
3. The "who else" question, on every call. "Who else needs to agree with this?" Write the answers into the CRM.
4. The sponsor by call two. Ask to meet the economic buyer early, while it is still a normal request.
5. A phase one that pays back alone. Value it in money and in the time the buyer's team will spend. Offer a checkpoint after phase one before they ask.
6. A forwardable one-page case. The champion must be able to send it without you in the room, with the two or three questions each role will ask already answered.

The realism line: you will not meet all 13 people. The champion has to carry the story to the rest, so make it easy to carry.

Where do multi-stakeholder deals leak?

Multi-stakeholder deals rarely die at the demo. They leak at quieter moments, and each leak is a skipped conversation:

None of these is fixed by a better document. Each needs a named call.

What does the buyer count that your value case misses?

The buyer counts their team's time. Late in a deal, pushback is often "make it simpler and more relevant", not "lower the price". The value case counts license and services. The buyer counts hours their people spend away from their jobs. The value case shows payback over the full program. The buyer wants to know whether phase one pays back alone, because that is the only promise a sponsor can defend without you. Price phase one on its own, in time as well as money. Some value only appears after the full sequence. Say so, and still price the checkpoint.

When can an AI agent sell and when does it take a rep?

An AI agent can handle the sale when one person buys and already knows the problem. It needs a rep when many people must agree on a vague problem. Two questions sort deals:

Plot your last ten wins on this grid. Most large deals sit in the bottom-right box. That work does not move to a chatbot when ten people have to agree.

Example: mapping one deal

Example: a seller's largest open deal is worth $400,000. The CRM shows four contacts. On the next call, the seller asks the champion who else needs to agree. The answer adds eight names across finance, IT, operations and procurement. The seller has met three of the 12. Two are known to doubt the problem exists.

The seller makes three moves. First, a one-page problem statement with the cost: $90,000 a quarter in lost stage-2 conversions, from the buyer's own funnel. Second, a 30-minute review of that page with the champion and the finance lead. Third, a phase-one plan that pays back in two quarters on its own. Coverage rises to 6 of 12 in three weeks. The deal does not close faster yet, but the conflict is now visible and named.

What should enablement build for buying groups?

Enablement should build for the group, not the persona. In practice:

• A problem-statement template and a one-page case template reps can fill in an hour.
• AI roleplays with two or three buyer personas in one meeting, so reps practice handling disagreement.
• A skills scorecard behavior: "confirms who else must agree".
• A CRM field for stakeholders met versus stakeholders named.
• A coaching question for managers: "Who in this deal have you never spoken to?"

See problem-first selling for how to open the conversation that consensus is built on.

Common mistakes

• Optimizing for the champion alone. The champion's enthusiasm does not survive a meeting you are not in.
• Personalizing for each stakeholder separately. It can split the group. Build one shared story.
• Treating multithreading as consensus. Contacts are not agreement.
• Emailing the proposal. Walk it through on a call where you can hear objections.
• Valuing only the full program. Price phase one on its own, including the buyer's time.

If you want to know whether your team has the hours to build and coach this, start with a free Capacity Check.