Buying Group Consensus: Why Arming One Champion Is the Wrong Job

The deal looked safe. Your main contact answered every message, sat through the technical review twice and told you, sincerely, that they would get this done. Then it went quiet, and the forecast note started saying internal alignment.

A decade of sales training has taught us to find one person inside an account and arm them for a fight. The evidence on how groups actually buy suggests that is the wrong unit of work. The committee is not waiting to be persuaded by somebody. It is trying to stop arguing with itself.

The Stall Has a Name

In fieldwork running from late summer into the autumn of 2024, Gartner put questions to 632 people on B2B buying teams. It reported that 74% of buyer teams show what it calls unhealthy conflict during the decision. The definition it uses is specific: members who hold incompatible objectives, who cannot agree on the right course of action, or who find themselves overruled by someone outside the group entirely. That is three quarters of your pipeline, carrying a disagreement you are not in the room for.

The flip side is in the same release. Groups that do reach agreement are 2.5 times more likely to call the deal a high-quality one, and buyers who experience what Gartner terms buying group relevance are three times more likely to say the same, according to the figures carried in coverage of the release. Gartner's own site refuses automated access, so treat those numbers as reported second-hand from the release and its syndicated copies rather than read off the research itself.

Delainey Kirkwood, a principal in Gartner's sales practice, draws the operational conclusion in the release: fostering agreement and reducing conflict has to be a priority for sales leadership, not a thing reps improvise. Kirkwood also notes the groups now run from five to sixteen people across as many as four functions, which is more opinions than any one champion can reconcile in a corridor conversation.

Why Better Champion Enablement Makes It Worse

The battle card. Every artefact we hand a champion is built for one person to win an argument: the deck, the comparison grid, the value calculator. They arm an individual position. The finance lead is not evaluating your argument, they are evaluating their own exposure, and a sharper version of someone else's case does nothing about that.

The personalisation. This is the counterintuitive one. In the same Gartner survey, content tailored to the buying group as a whole lifted consensus by 20%, while content pitched at the individual hurt it by 59%. Kirkwood attributes the effect to confirmation bias: each stakeholder reads the version written for them, each becomes more certain, and the gaps between them widen rather than close.

The extra call. Adding people to a meeting without giving them a shared problem statement produces more positions, not a decision. Everyone leaves with their original view intact and a new sense that this is complicated.

Sellers Target People, Buyers Decide in Committee

The mismatch is structural. Every tool we own segments, sequences and scores an individual. The decision happens somewhere we cannot see, usually in a thread, and the question in that thread is not whether your product is good. It is whether these particular colleagues can agree on what they are buying without anybody losing face.

It is also a measurement gap. A CRM records contacts, stages and amounts. There is no field anywhere in it for whether finance and operations describe the problem the same way. What nobody records, nobody manages, which is the case our piece on mapping the committee before you need it makes from the opposite end.

Picture the usual version. Operations wants reporting that arrives before the meeting. Finance wants the total to stop moving. Security wants to know who can read what. Each of them receives a well-targeted message that confirms the thing they already believed. Nobody reads the other two. Three weeks later your contact tells you, truthfully, that they are working on it.

Run the Premortem Before the Group Runs It For You

The most useful import here comes from outside sales entirely. Gary Klein described the premortem in Harvard Business Review in September 2007: a team assumes the project has already failed and then explains why. Klein's reason for the inversion is that "too many people are reluctant to speak up about their reservations" while a plan still looks healthy, and the exercise makes it safe for the informed sceptic to talk.

Borrow it one stakeholder at a time. Ask each person separately what would make this a bad decision in twelve months. You will get answers nobody would volunteer in a group call, and the places where two answers contradict each other are the actual obstacle in the deal. Then do the thing that feels risky: put the contradiction on the table as a design question rather than a problem. Finance needs the number to be predictable and operations needs it to be fast; here are two shapes that do both. That is a sentence a group can work with. Our guide to running discovery across several stakeholders covers how to get those separate conversations booked without it looking like you are going around anyone.

One Document Everybody Reads

Replace one personalised email per deal with a single shared plan written in neutral language, which the champion can forward without editing. The Dock team's library entry on mutual action plans, published in March 2025 and updated this year, gives a sensible skeleton: an overview with two or three goals, the success criteria in the buyer's own metrics, the named people and their roles on both sides, and a timeline broken into phases. It also advises keeping the whole thing to "the most important 8-12 milestones", which is the discipline most of these documents lack.

One further detail from the same piece is worth stealing. Anchor the dates to the buyer's go-live rather than your close date: "If the client's go-live date is set for July 1st, start there and work backward". A plan built backward from a date the buyer chose is a plan the group can argue about on its merits. A plan built forward from your quarter end is a plan with your name on it, and groups do not reach consensus around a vendor's calendar.

Personalisation Is the First Thing to Cut

Here is the position, and it is deliberately uncomfortable: in a multi-stakeholder deal, most individual-level personalisation is actively harmful, and the right move is to spend that effort on one group-facing artefact instead. The Gartner figures point the same direction, and the mechanism Kirkwood names is the reason to believe it rather than just the correlation. Confirmation bias is not a vendor theory, it is one of the better-replicated findings in decision research, and every message written to make one stakeholder nod harder is a small push away from agreement.

This argues against a lot of current tooling. Sequencing engines, intent-triggered messaging and research assistants are all optimised to produce more individually relevant touches, faster. They will do that very well, and in a committee deal the output is a more confident set of disagreeing people.

A small buying group that already agrees is the opposite case. Where one or two people hold the decision and the problem is simply getting attention, individual relevance is exactly right and a group document is overhead nobody asked for. Use the committee's own temperature to decide. If your separate premortem conversations come back with broadly the same answer, personalise freely and move fast. If they come back contradicting each other, stop writing to individuals and start writing to the room, and read our note on how fit and timing have to be judged together before you decide how much more to invest.

Put It in the Deal Review

• Add one field to the review: do the stakeholders describe the problem the same way? Yes, no, or not yet known. Not yet known is the most honest answer most weeks.
• Give every enterprise deal one forwardable document, and count how often it actually gets forwarded.
• Make the premortem question standard in every multi-stakeholder deal, asked one to one.
• When a deal stalls, record which disagreement caused it rather than a stage name. Stalls that end in nothing have their own pattern, which our piece on deals lost to no decision breaks down.

Expect the forecast to look worse before it looks better. Finding the disagreement early means admitting it exists while there is still time to design around it, and a quarter that gets more honest in week four is easier to defend than one that discovers the truth in week eleven. Outcomes vary by deal, by market and by how much room the group has to move, and no process guarantees anything. Pick one open deal this week and write the single paragraph describing the problem that every stakeholder would sign. If you cannot write it, you have found your next conversation.